JP2010281781A - Driving lane detector - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide "a lane change detector" that detects a present driving lane by an image of a backsight camera, and does not cause detection delay even when the image of the backsight camera is used, because, even when the lane is changed, it appears on the image of the backsight camera with time lag and detection delay of the lane change is caused. <P>SOLUTION: When the driving lane is detected by the pick up image of the backsight camera, the tilt angle of a lane boundary line is measured. When tilting of the lane boundary line by a predetermined angle or more is measured and detected, the moving distance of the lateral direction of a vehicle is calculated based on the measured tilt angle, the vehicle speed, and the elapsed time from the start time of the measurement. When the lateral direction movement amount is a predetermined distance or more, it is determined that the vehicle changes the lane even when the movement of the lane is not detected by the backsight camera, and the display of the travel lane is changed. At this time, when data such as increase of the lanes by a right-turn lane is previously inputted from a database, it is determined that the driving lane is changed if the lateral movement occurs with that timing, and early read-ahead is allowed. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a traveling lane detection device that can quickly and accurately detect which lane a vehicle is currently traveling when traveling on a road with multiple lanes on one side and changing lanes.

  2. Description of the Related Art A navigation device mounted on a vehicle includes a map / information data recording medium such as a DVD or a hard disk on which map data for drawing a map and facility information data for displaying and retrieving facilities are displayed. . The navigation device uses a data capture device that captures data from this map / information data recording medium, a monitor that displays a map, etc., and a self-contained navigation device that uses a GPS receiver, a mileage sensor, a gyroscope, etc., if necessary. Has a current position detection device that detects the current position and the direction of travel, fetches map data including the current position from a map / information data recording medium, and monitors a map image around the current position based on the map data The current position mark is superimposed on the monitor screen and displayed on the monitor screen, the map image is scrolled as the vehicle moves, or the map image is fixed on the screen and the current position mark is moved. You can see at a glance where you are moving.

  Although recent navigation devices have improved the accuracy of current position data obtained by GPS, it cannot always be said that it has reached a practical level to quickly and accurately detect the lane in which the vehicle is currently traveling. On the other hand, it is desired to provide guidance so that the lane can be changed with a margin in front of an intersection that should turn right or left.

  On the other hand, recent vehicles are equipped with a camera that captures the outside of the vehicle for the safety of vehicle travel, and the captured image is displayed on a monitor. Systems that prevent accidents with pedestrians and sometimes do not come into contact with dangerous goods are widespread.

  In recent years, cameras that shoot outside the vehicle have become cheaper. In addition to the camera that shoots from the rear as described above, a camera that shoots the front and sides is installed, and the images taken by each camera are displayed on the monitor by switching the cameras appropriately. There are some that can be displayed. Furthermore, with the top view system, the captured images of these cameras are image-processed, and the surroundings of the vehicle are displayed on the monitor in an easy-to-understand manner as if the surroundings of the vehicle were photographed with a single camera installed above the vehicle. There is also a system called.

  Furthermore, for the camera that captures the front of the vehicle, for safety when the vehicle is about to enter an intersection with poor visibility, a camera that captures the left and right of the front of the vehicle is used. It is also being used for applications such as maintaining distance, and it is envisioned that these systems will become more popular.

  There has also been proposed a technique for detecting the driving lane on the road on which the vehicle is currently traveling and providing information to the driver using such a camera that captures the outside of the vehicle, and in particular, using a camera that captures the front of the vehicle. Patent Documents 1 to 3 below disclose techniques that recognize a lane that is currently being traveled and that can be used to guide a navigation device.

Japanese Patent Laid-Open No. 10-300494 JP 2005-345240 A JP 2001-82975 A

  Although techniques have been proposed to recognize and guide the lane that is currently running using the image of the camera that captures the outside of the vehicle as described above, they are techniques that use a camera that captures the front of the vehicle, It is a camera that is not often used for other purposes. That is, for example, a camera that captures the rear of the vehicle captures an area that is most difficult to see for the driver and can be used very effectively when parking or the like, whereas it captures the front of the vehicle as described above. The camera is used for a special purpose and can be said to be a camera that is less necessary for many people. Therefore, when trying to recognize the lane using such a camera that captures the front, in many cases, a camera is newly established only for that purpose. It must be expensive.

  Therefore, if it is possible to recognize the currently traveling lane using a widely used rear-view camera, there is no need to use a camera for photographing a special front. However, when the rear-view camera changes the lane, the rear-view camera detects that the lane has changed after all the vehicles have moved to the adjacent lane. That is, for example, when driving on a three-lane road on one side as shown in FIG. 7 (a), when trying to make a right turn at an intersection to travel along the guidance route, from the position of L4c in front of the intersection When a right turn lane exists, the driver enters the right turn lane, which is the fourth lane, from this position. When such a run is performed, an image of a rear-view camera provided behind the vehicle is obtained as an image shown in C1 of FIG. Furthermore, by taking in the data of the lane near the current position from the database, it is a road that has three lanes on one side, and has been running for a while from the image recognition of the rear-view camera by running for a while. Confirmation of the number of lanes on the road, and it is detected that the lane in which the host vehicle is traveling is in the third lane.

  After that, even if the vehicle moves to the fourth lane that is the right turn lane, it is recognized that L4c that is the tip of the lane boundary line L4 that separates the third lane and the fourth lane that is the right turn lane is not captured. Therefore, it cannot be recognized that the vehicle has entered the right turn lane as indicated by C2 in FIGS. After that, it can be detected from the image of C3 that the vehicle has moved to the fourth lane, which is the right turn lane, for the first time.

  Therefore, in a vehicle navigation system equipped with a rear-view camera, it is detected from the database that the vehicle is traveling on a three-lane road, and it is confirmed that the vehicle is traveling in the third lane by image processing of the rear-view camera. When the vehicle enters the right turn lane, which is the fourth lane, through the guidance area to the right turn lane, it is immediately detected from the database that there are four lanes. Since the rear-view camera cannot detect the lane boundary line L4 between the third lane and the fourth lane depending on the image of C2, it is determined that the vehicle is still traveling on the third lane of the four-lane road. At the same time, the own vehicle position mark traveling in the third lane is displayed on the monitor screen. Such a state is also largely due to the presence of a wide guidance area U having no lane boundary line at the introduction to the right turn lane as described above.

  As a result, the navigation apparatus outputs a voice guidance or the like for prompting attention to move to the third lane as soon as possible. At this time, the driver hears the guidance that he should enter the right turn lane even though he is already in the right turn lane. It can be thought that there is.

  In addition, the problem of the recognition delay of the traveling lane when detecting the currently traveling lane with such a rear-view camera is that the number of lanes is detected in the database as described above, and the driver Even if the lane has already been changed, the rear-view camera still does not recognize this, which causes a significant problem.In addition to this, for example, when moving to the left lane to turn left at an intersection, The recognition by the image taken by the visual camera will detect that the lane has changed for the first time when the lane is almost completely changed, so the lane should be changed for the driver who has almost completed the lane change. A similar problem occurs.

  Therefore, the lane change detection device according to the present invention is a system that detects a lane that is currently running from images of a rear-view camera mounted on many vehicles, and that the vehicle is running almost completely changing lanes. The purpose is to solve the problem of the lane change detection delay that detects the lane change for the first time when detecting the lane change, especially the database detects that the number of lanes such as right turn lanes has increased, and the vehicle has already traveled on that lane In spite of this, an object is to solve the problem of determining that the vehicle is still traveling in another lane due to a delay in recognition of the lane by the rear-view camera.

  In order to solve the above-described problem, the lane change detection apparatus according to the present invention measures a lane boundary line inclination angle measurement unit based on a captured image of a rear-view camera, and the lane boundary line inclination angle measurement unit. Lateral movement that detects the distance the vehicle has moved in the lateral direction based on the measured inclination angle, vehicle speed, and elapsed time from the start of the measurement. An amount detection unit, and a lane change detection unit that detects that the vehicle has changed lanes when the lateral movement amount detected by the lateral movement amount detection unit is equal to or greater than a predetermined distance.

  The lane change detection device according to the present invention includes a lane number change detection unit that detects a change in the number of lanes of a road that is currently running from a database in the lane change detection device, and the lane number change detection unit includes: When it is detected that the number of lanes changes within a predetermined distance, the lane change detection unit detects a lane change due to a change in the number of lanes when the lateral movement amount detection unit detects that the vehicle has moved laterally more than a predetermined distance. It is characterized that it is determined that a change has been made.

  In the lane change detection device according to the present invention, in the lane change detection device, the lane change detection unit detects that the lane number changes when the lane number change detection unit detects that the lane number changes. It is characterized in that it is determined that a lane has been taken when the amount of lateral movement is smaller than in the case where the vehicle does not.

  In the lane change detection device according to the present invention, in the lane change detection device, the traveling lane detection unit detects a lane in which the vehicle is currently traveling from a photographed image of the rear-view camera during steady traveling, and the lane When the lane change is detected by the change detection unit, it is determined that the lane change has been made even when the lane change is not displayed on the captured image of the rear-view camera.

  The lane change detection device according to the present invention is characterized in that, in the lane change detection device, the rear-view camera is a camera that captures the rear of the vehicle when the vehicle moves backward.

  Since the present invention is configured as described above, it does not use a camera that shoots the front, which is rarely used, and is currently traveling using images of a rear-view camera mounted on many vehicles. In the system for detecting the lane, it is possible to solve the problem of the detection delay of the traveling lane due to the photographed image of the rear-view camera being photographed after the vehicle moves.

  In particular, the database detects that the number of lanes such as a right turn lane has increased, and the vehicle is already driving in that lane, but it is still in another lane due to a delay in recognition of the driving lane by the rear-view camera. Therefore, it is possible to solve the problem of performing guidance for prompting the driver to pay attention by judging that the vehicle is traveling.

It is a functional block diagram of the Example of this invention. It is an operation | movement flowchart which shows the basic operation | movement of the Example. It is an operation | movement flowchart which performs the process of step S5 of FIG. It is an operation | movement flowchart which performs the process of step S6 of FIG. It is a figure which shows the driving | running | working state when a right turn lane increases during driving | running | working on the one side 3 lane road, and a vehicle moves to a right turn lane, and the example of a picked-up image of a backsight camera. It is a figure which shows the driving | running | working state which changes a lane for a left turn during driving | running | working on the one-side 2 lane road, and the example of a picked-up image of a backsight camera. It is a figure which shows the other example of the driving | running | working state when a right turn lane increases during driving | running | working on the one-side three lane road, and a vehicle moves to a right turn lane, and the picked-up image of a backsight camera.

  Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram of an embodiment of the present invention, in which an image captured by a rear-view camera provided at the rear of a vehicle is captured in this system by a captured image capturing unit 2. This rear-view camera 1 detects whether there is no dangerous object behind so that it can be surely entered into the parking space, for example, when entering the parking space in the back, which is currently installed in many vehicles. Alternatively, it is possible to use a rear-view camera that is mainly used for checking whether there are children or pedestrians.

  The images of the rear-view camera captured by the captured image capturing unit 2 are used in various functional units, and in the illustrated example, are output to the lane boundary line inclination measuring unit 3 in order to implement the present invention. In this lane boundary line inclination angle measuring unit 5, the angle at which the lane boundary line for displaying the boundary line between the lanes existing in the image of the captured image capturing unit 2 is inclined with respect to the traveling direction of the vehicle. taking measurement.

  That is, for example, in the example shown in FIG. 5, when driving on the third lane of the three-lane road first, when turning to the fourth lane of the right turn lane, which is an increasing lane, to turn right at the intersection When the vehicle turns to the right to enter the right turn lane and the vehicle rotates as shown, the rear-view camera image changes from the state A1 to the state A2 in FIG. 5B. At this time, the lane boundary line displayed in the image of the rear-view camera is inclined by the angle α. Therefore, the lane boundary line inclination angle measuring unit 3 measures this angle to obtain the inclination angle of the lane boundary line, and by measuring the angle, a substantial vehicle snake angle can be detected.

  The captured image upper lane number detection unit 4 detects the lane boundary line from the captured image of the rear-view camera, and detects the number of lanes on the currently traveling road. This detection of the number of lanes has the same function as that in the case where the number of lanes on the road that is currently running is detected from the database 10 based on the current position data of the current position detection unit 9 in the database upper lane number detection unit 11 described later. However, this function is not necessarily required, but since the current position data is required in the database lane number detection unit 11, the current position data is obtained when the vehicle has traveled through the group of buildings for a relatively long time. In such a case, the number of lanes obtained from this captured image can be used to detect the number of lanes more reliably.

  The lane boundary horizontal movement detection unit 5 often crosses the lane boundary line obliquely when the vehicle changes lanes, and the state is detected by a captured image of the rear-view camera. At the time of the detection, for example, when the position of the host vehicle that is always present at the center in the captured image intersects the lane boundary line that is inclined and moving in the lateral direction, it can be detected that the vehicle has moved across the lane boundary line. . Eventually, it is possible to detect the currently traveling lane by confirming in which lane position of the lane boundary line of the entire rear-view camera photographed image the vehicle is present.

  When changing lanes, it does not necessarily cross the lane boundary line, and when the right turn lane increases as will be described later, it enters the right turn lane without crossing the lane boundary line, and it is completely captured by the image taken by the rear-view camera. Since the vehicle must travel until the number of lanes is confirmed, the travel lane detection unit 13 (to be described later) changes the travel lane in a manner that does not depend on the number of lane boundary lines photographed by the rear-view camera. Detection of movement is required.

  The lateral movement amount detection unit 8 in FIG. 1 detects a vehicle speed based on the lane boundary line inclination angle detected by the lane boundary line inclination angle measurement unit 3 and vehicle speed data used for, for example, vehicle speed display. Based on the signal of the detection unit 7 and the elapsed time data from the time when the lane boundary line inclination angle detection unit from the elapsed time measurement unit 6 detects that the lane boundary is inclined, the amount of movement of the vehicle in the lateral direction is calculated. Integrate and detect. That is, when the vehicle rotates the steering wheel to the right or left to change the lane, the angle of the vehicle changes, and when the angle α of the lane boundary line in the captured image of the rear-view camera gradually changes, By integrating the angle and the amount of movement, it is possible to detect the amount of lateral movement since the vehicle moved in the lateral direction.

  The traveling lane detection unit 13 detects a lane that is currently traveling, and normally uses the captured image of the rear-view camera, and the lane number of the road currently traveling by the captured image upper lane number detection unit 4. The number of lanes of the currently traveling road is determined based on the number of lanes of the currently traveling road detected by the lane number detection unit 11 in the database as necessary. From the image, the currently running lane portion in the lane is detected. The travel lane detection unit 13 includes a travel lane change detection unit 14 that detects that the travel lane has been changed, and particularly detects that the travel lane has been changed when detecting the travel lane.

  Data on the lane number detection unit 11 in the database is input to the travel lane detection unit 13, and the database lane number detection unit 11 is based on the data of the current position detection unit 9 and includes a DVD or the like included in the navigation device. The number of lanes on the road that is currently running is read from a database recorded on a map / information data recording medium such as a hard disk, and this is output to the running lane detector 13.

  The database upper lane number detection unit 11 detects the number of lanes together with the photographed image upper lane number detection unit 4, but the lane number detection method is significantly different from the photographed image upper lane number detection unit 4, respectively. Since it has advantages and disadvantages, the traveling lane detector 13 and the like can properly use them as necessary. Therefore, each lane number detection data can be used for detecting the lane number change, but in the present invention, in particular, in order to be able to respond quickly to an increase in the number of lanes due to the presence of a right turn lane, Detection is mainly used.

  The travel lane detection unit 13 also inputs data of the database upper lane number change detection unit 12. In the database upper lane number change detection unit 12, the number of lanes detected by the database upper lane number detection unit 11 is, for example, A change in the number of lanes, such as an increase from 3 lanes to 4 lanes, is detected and output to the traveling lane position detection unit 12.

  The travel lane change detection unit 14 in the travel lane detection unit 13 includes a travel lane change detection unit 15 at the time of change in the number of lanes on the database. When it is detected that the number of lanes changes due to the presence of a lane or the like, a change in the traveling lane is detected.

  In such a case, the number of lanes changes in many cases by providing a right turn lane from the front of the intersection, and as described in the subject of the present invention, the traveling lane is simply detected by the image captured by the rear-view camera. Prevent inappropriate navigation device guidance. The travel lane change detection process at this time can be implemented by an operation flow of FIGS. 2 and 3 to be described later and an aspect of FIG.

  In the illustrated example, the travel lane change detection unit 14 is provided with a travel lane change detection unit 16 during steady travel. Here, it is assumed that the travel lane change detection when the number of lanes is changed is not during steady travel, and other travel times are detected. At the time of steady running, lane change detection processing at that time is performed. Therefore, the switching of the operation of the travel lane change detection unit 15 at the time of changing the number of lanes on the database and the operation of the change lane change detection unit 16 at the time of steady travel are switched by the input of a signal with the lane number changed from the database lane number change detection unit 12. be able to.

  The traveling lane detection unit 13 outputs the number of lanes on the currently traveling road and the detection result of the currently traveling lane to the map display processing unit displaying the map from the current position display unit on the traveling lane, Overlay the map screen. When map data including a lane in the vicinity of a road to be traveled is already read from the database and displayed in the map display processing unit, the current position is displayed on the lane displayed there.

  The traveling lane detection device including the functional blocks shown in FIG. 1 can be implemented by operating in the order according to the operation flow shown in FIGS. The operation flow shown in FIGS. 2 to 4 will be described below with reference to the functional block diagram of FIG. 1 and the operation examples of FIGS. FIG. 2 shows the overall operation flow of the present invention. In the rear-view camera image using lane change detection process in FIG. 2, the number of lanes on the road currently running from the first database or the rear-view camera image is shown. Is detected (step S1).

  This operation is constantly performed when the vehicle is running and the navigation device is operating. First, the database 10 has the current position detected by the current position detecting unit 9 in the database upper lane number detecting unit 11 of FIG. 1 and by reading the number of lanes on the road that is currently running, and secondly, it can also be detected as described above by the on-capture lane number detection unit 4 in FIG. As described above, it is possible to select data as needed, for example, using data in the database and using an image captured by the backsight camera when the current position cannot be detected.

  Next, the traveling lane is recognized by the rear view camera (step S2). In this state, in the road plan view of FIG. 5 (a), an image is captured by the rear-view camera at the vehicle position P1, and an image in the range of the image A1 is processed from the image, and an example of the captured image of FIG. An image indicated by A1 can be input. Although the illustrated image can be further widened, a narrow range is shown for comparison in one drawing. It can be seen from the image A1 in FIG. 5 (b) that the vehicle is currently traveling on a three-lane road and is traveling on the third lane on the center line CL side. The center position of the host vehicle exists at the center of the image, and a total of three lanes are displayed in parallel on the left side in the traveling direction of the vehicle.

  Thereafter, the current number of road lanes and the travel lane are displayed based on the data in steps S1 and S2 (step S3). This display is a monitor that displays a map screen of the navigation device. A road screen corresponding to the number of lanes is displayed on the map screen, and the vehicle position mark is displayed on the road lane recognized in step S2 on the road screen. This can be done.

  Next, in the embodiment shown in FIG. 2, it is determined whether or not the number of lanes on the road that is currently being read from the database changes within a predetermined distance (step S4). That is, as described above, the method of detecting the traveling lane by the rear-view camera is effective because the rear-view camera is extremely versatile for parking and has no speciality like the front camera for photographing the front. The timing to detect that the driving lane has changed is slower than when using the image of the front camera, and it is much slower than detecting the number of lanes on the road on which the vehicle is currently running, The main purpose is to solve the problems that hinder the guidance of the navigation device, so the number of currently running lanes that are currently read from the database changes whether or not this is the case. Whether or not is determined by the data.

  When it is determined in step S4 that the number of lanes on the road that is currently being read from the database changes within a predetermined distance such as 10 m from the position, the process proceeds to step S5, and when the change in the number of lanes shown in FIG. 3 is detected. The travel lane change detection process is performed. On the other hand, when it is determined that the number of currently running lanes read from the database does not change within a predetermined distance, the process proceeds to step S6, and the number of lanes when no change in the number of lanes is detected as shown in FIG. Perform change detection processing. After the processes of step S5 and step S6 are performed, the process returns to step S1 and the operation is repeated.

  In the guidance route preset by the navigation device, it is preferable to change lanes soon when turning right or left at this intersection, and when there is a right turn lane, enter that lane. Etc. is transmitted to the driver by voice or the like. Further, when approaching an intersection and not yet being able to cope with the detection of the traveling lane, guidance is provided for moving to the right turn lane or the left turn lane.

  In the travel lane change detection process at the time of detecting the change in the number of lanes in step S5, as shown in FIG. 3, it is first determined whether or not the lane is inclined in the image of the rear-view camera (step S11). Here, when it is determined that the lane is not inclined in the image of the rear-view camera, the process proceeds to step S18 and then returns to step S1 in FIG. 2 to repeat the above operation. On the other hand, when it is determined that the vehicle is inclined, the inclination direction and the inclination angle are measured (step S12).

  In the example of FIG. 5, an image captured by the rear-view camera at the vehicle position P2 in FIG. 5 (a) is obtained as A2, and this is shown in an enlarged view of the image A2 in FIG. 5 (b). It is detected that all the lane boundary lines are inclined in parallel by the angle α1 with respect to the traveling direction of the vehicle. Therefore, when such a tilt is detected in step S11, it is determined that the driver has operated the steering wheel to change the traveling direction of the vehicle, and the tilt direction and tilt angle are measured (step S12).

  Thereafter, the amount of movement in the lateral direction is calculated from the inclination angle of the lane and the travel distance (step S13), and it is determined whether or not the first predetermined amount has been moved in the lateral direction based on the calculation result (step S14). When it is determined that the first predetermined amount such as 1.5 m has not been moved, it is determined whether or not the vehicle has returned to the original travel position in the example shown in FIG. 2 (step S17). Returning to step S12, the operation is repeated, and the accumulation of the movement amount in the lateral direction is continued. When it is determined in step S17 that the vehicle has returned to the original travel lane, it is determined that the change in the traveling direction of the previous series of vehicles was temporary, the process proceeds to step S18, and the process returns to step S1 in FIG. Repeat operation.

  In the example shown in FIG. 5, a steering operation is performed to change the direction further to the right at the vehicle position P3 in FIG. 5A. In the image of the rear-view camera at that time, the angle α3 and its angle are as shown in the image A3. Largely, it can be determined that the vehicle is changing lanes by changing the inclination angle of the lane up to that point and integrating the travel distance.

  At this time, even if the first predetermined amount is relatively small, such as 1 m or 1.5 m, for example, when the data on the number of lanes to be increased is obtained from the database, at that timing, It can be estimated that the vehicle has moved in the lateral direction because it is due to a lane change, so even if there are relatively few lateral movements, it is assumed that the lane has been changed, and the vehicle moves to the lane side where the vehicle position changes early. You may do it.

  In the operation flow shown in FIG. 3, in the process of accumulating the amount of movement in the lateral direction from the inclination angle of the lane boundary line and the travel distance of the vehicle, it moves in the lateral direction for a first predetermined distance such as 1.5 m, for example. If it is determined that the vehicle lane number has changed, it is estimated that the lane change has occurred due to a change in the number of lanes (step S15), and the vehicle position mark is moved and displayed on the changed diagonal line (step S16).

  Thus, conventionally, for example, even at the vehicle position P4 in FIG. 5 (a), it has not been possible to detect the increase in the number of lanes from the captured image A4 of the rear view camera. In contrast, in the above embodiment, for example, the vehicle position mark can be moved to the fourth lane and displayed even at the vehicle position P3. Also, when the navigation device should turn right at this intersection on a preset guidance route, the driver can quickly detect that the operator is turning right at this intersection. It is possible to avoid unnecessary guidance such as “Please move to the right turn lane”. The above operation is performed by the travel lane change detection unit 15 when the number of lanes on the database in FIG. 1 is changed.

  On the other hand, in the traveling lane change detection process at the time of non-detection of the change in the number of lanes of FIG. 4 which is the process of step S6 of FIG. 2, an example is shown in which the operation substantially similar to the operation of FIG. Since S21 to S23 are the same, detailed description is omitted, but in this process, it is determined that the number of currently running lanes read from the database does not change within a predetermined distance as described above. This processing is a routine lane change detection process using a rear-view camera. This processing is carried out by the traveling lane change detection unit 16 during steady driving shown in FIG.

  In the process shown in FIG. 4, it is first determined whether or not the lane is inclined based on the image of the rear-view camera (step S21). When the lane is inclined, the inclination direction and the inclination angle are measured (step S22). Then, the amount of movement in the lateral direction is calculated from the travel distance (step S23), and then it is determined whether or not the second predetermined amount has been moved in the lateral direction (step S24). The second predetermined amount here is preferably larger than the first predetermined amount in step S14 of FIG.

  That is, in the operation of FIG. 3, when the information that the number of lanes changes within a predetermined distance such as 10 m from the current position is obtained in advance by the database, the probability of changing the lane at this position is quite high. Since it is estimated, it can be estimated that the lane has been changed even by a lateral movement of a relatively small distance. However, when such information is not available from the database, it is estimated that the lane has been changed to be the same as or slightly less than the general lane width. Therefore, the second predetermined distance at this time is set to be larger than the first predetermined distance, and it is determined that the lane is changed by a large lateral movement.

  When it is determined in step S24 that the second predetermined amount such as 2 m has not been moved, the process proceeds to step S27 to determine whether or not the vehicle has returned to the original travel position, and when it has been determined that the vehicle has returned to the original travel position. Proceeding to step S28, the process returns to step S1 of FIG. If it is determined in step S27 that the vehicle has not returned to the original travel position, the process returns to step S22 and the lateral movement amount accumulation process is continued.

  As a result of such an operation, for example, as shown in FIG. 6, when traveling on the second lane of the two-lane road at the vehicle position P1 in FIG. As a result of turning the steering wheel, the image captured by the rear-view camera at the vehicle position P2 changes from the image B1 to B2 in FIG. 5B, and the lane boundary line is inclined by β1.

  In this state, it is determined that the number of currently running lanes read from the database in step S4 in FIG. 2 does not change, and then in step S21, it is determined that the lane is inclined in the image of the rear-view camera, and in step S22. This corresponds to the state where the lane direction and the inclination angle are measured. Here, it is detected that the lane is relatively rotated rightward by β1 as the direction of the vehicle changes to the left.

  Thereafter, in step S23, the lateral movement amount is calculated from the lane inclination angle and the travel distance, and when the vehicle moves further from the vehicle position P3 in FIG. 6A to P4, B3, B4 and the image from the image in FIG. In this process, it is determined in step S24 that the second predetermined amount has been moved in the lateral direction. When it is determined in step S14 that the vehicle has moved, for example, a predetermined distance of 2 m laterally, the process proceeds to step S25, where the travel lane position is changed to the adjacent lane, and the operation returns to step S1 in FIG. repeat.

  As a result, even if the vehicle position P3 in FIG. 6 (a) does not detect that the vehicle has moved to the adjacent lane in the rear-view camera image, as in the image B3 in FIG. When it is detected that the vehicle has moved, for example, 2 m, which is a predetermined distance of 2, the current vehicle position displayed on the monitor can be moved to the first lane. Therefore, before the lane change determination based on the image captured by the rear-view camera detects that the lane has been changed by the image B4 in FIG. 6B at the vehicle position P4 in FIG. 6A, the vehicle changes the lane. The vehicle travel position can be moved by estimating what has been done.

  In the example shown in FIG. 1, the lane boundary crossing movement detection unit 4 is provided as described above, and the center line of the vehicle crosses the lane boundary in the process of changing from the image B3 to B4 in FIG. Therefore, it is possible to detect that the vehicle has moved to the next lane at this moment. By this method as well as by using this method in combination as necessary, more accurate lane change detection can be performed.

  By performing the processing as described above, the navigation device does not have a front camera for photographing the front of the vehicle, and the navigation device turns left even if the captured image of a commonly used rear-view camera is used. When instructing to change lanes, it is assumed that the vehicle has not changed lanes just before the intersection, and there is no need to promptly change to the left lane. It won't cause any confusion in driving operations.

DESCRIPTION OF SYMBOLS 1 Rear view camera 2 Captured image capture part 3 Lane boundary line inclination angle measurement part 4 Captured image upper lane number detection part 5 Lane boundary line lateral movement detection part 6 Elapsed elapsed time measurement part 7 Vehicle speed detection part 8 Lateral movement amount detection Section 9 Current position detection section 10 Database 11 Database upper lane number detection section 12 Database upper lane number change detection section 13 Travel lane detection section 14 Travel lane change detection section 15 Travel lane change detection section 16 when database upper lane number changes Traveling lane change detection unit 17 Current position display unit on traveling lane

Claims (5)

A lane boundary inclination measuring unit that measures an inclination angle of the lane boundary line from an image captured by the rear-view camera;
When the lane boundary inclination angle measurement unit measures that the lane boundary line is inclined more than a predetermined amount, the vehicle moves in the lateral direction according to the measured inclination angle, the vehicle speed, and the elapsed time from the start of the measurement. A lateral movement amount detection unit for detecting the measured distance;
A lane change detection device comprising: a lane change detection unit that detects that the vehicle has changed lanes when the lateral movement amount detected by the lateral direction movement amount detection unit is equal to or greater than a predetermined distance. A lane number change detection unit that detects a change in the number of lanes on the currently running road from the database,
When the lane number change detection unit detects that the number of lanes changes within a predetermined distance, the lane change detection unit detects when the lateral movement amount detection unit detects that the vehicle has moved laterally by a predetermined distance or more. The lane change detection device according to claim 1, wherein it is determined that a lane change due to a change in the number of lanes has been performed.   When the lane number change detection unit detects that the lane number changes, the lane change detection unit determines that the lane change has been performed when the lateral movement amount is smaller than when the lane number change is not detected. The lane change detection device according to claim 1.   The traveling lane detection unit detects a lane in which the vehicle is currently traveling from a captured image of the rear-view camera during steady traveling, and a captured image of the rear-view camera when the lane change detection unit detects a lane change. The lane change detection device according to claim 1, wherein it is determined that the lane change has been made even when the lane change is not displayed.   The lane change detection device according to claim 1, wherein the rear-view camera is a camera that captures the rear of the vehicle when the vehicle is moving backward.

Images