JP2011128049A - Traffic sign recognition device and traffic sign recognition method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a traffic sign recognition device and a traffic sign recognition method capable of determining accurately and highly speedily whether a lane on which a vehicle travels is a specific lane. <P>SOLUTION: A traffic sign recognition device includes a camera, a lane mark detection unit for detecting a lane mark on the basis of an image photographed by the camera, a specific traffic sign detection unit for detecting a specific traffic sign on the basis of an image photographed by the camera, a travel distance calculation unit for calculating a travel distance of an own vehicle, and a specific traffic sign presence determination unit for determining whether the own vehicle travels on a specific lane when a specific traffic sign is detected on a lane on which the own vehicle travels before the vehicle travels a predetermined distance. The lane mark detection unit identifies a right-side lane mark and a left-side lane mark relative to the lane on which the own vehicle travels, and the specific traffic sign presence unit determines that the own vehicle travels on the specific lane when a position of the specific traffic sign is located between a position of the right-side lane mark and a position of the left-side lane mark. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a road marking recognition apparatus and a road marking recognition method, and more particularly to a road marking recognition apparatus and a road marking recognition method that can accurately and quickly detect a mark marked on a traveling lane of a host vehicle.

  Conventional typical vehicle-mounted navigation devices include a control device such as a CPU that controls all processes related to navigation, a storage device such as a DVD (Digital Versatile Disk) -ROM or an IC memory card that stores map data in advance, It has a display device, a GPS (Global Positioning System) receiver, a detection device for detecting the current position and current direction of the vehicle, such as a gyroscope and a vehicle speed sensor. Then, the control device reads out map data including the current position of the vehicle from the storage device, displays a map image around the vehicle position on the display device screen based on the map data, and indicates the current position of the host vehicle. The vehicle position mark to be displayed is superimposed on the map image, the map image is scrolled according to the movement of the vehicle, the map image is fixed on the screen and the vehicle position mark is moved, and where the vehicle is currently located. You can see at a glance whether you are driving.

  The vehicle-mounted navigation device is usually equipped with a function (route guidance function) for guiding the user so that the user can easily travel on the road without making a mistake. According to this route guidance function, a simulation such as a horizontal search method or a Dijkstra method is used by the control device to search for an optimum route from the departure place (typically the current position of the vehicle) to the destination using map data. Performs automatic search by calculation, stores the searched route as a guide route, and makes it possible to distinguish the guide route from other roads on the map image while driving (for example, changing the color or line width) When the vehicle approaches a predetermined distance to the intersection where the course should be changed on the guidance route, a guide map of the intersection (enlarged view of the intersection, at the intersection) is displayed on the map image. By displaying an arrow indicating the traveling direction of the vehicle, the user can grasp which road should be traveled.

  In the search for such a guidance route, a search according to the traveling lane may be required. For example, a freeway in the United States has a road with a special lane called a car pool lane along with a general lane. Carpool lane is a lane established to eliminate traffic congestion, reduce exhaust gas, shorten commuting time, etc., and has two or more public buses and passengers (specific numbers vary by state) It can be used by vehicles and motorcycles. Such a car pool lane is not allowed to enter legally from a general lane unless it is from a predetermined entry permission section. In addition, a legal departure to a general lane is not permitted unless it is from a predetermined departure permission section.

  As a technique for searching for an appropriate guidance route according to such a special road form, Patent Literature 1 detects a road marking on a captured image obtained by an in-vehicle camera, and within a predetermined time, a car pool lane. It is determined whether or not the vehicle is traveling in the car pool lane by detecting the road marking of the vehicle, and when the vehicle is traveling in the car pool lane, information on the permitted entry section to the general lane is described. Has been.

Japanese Patent Laid-Open No. 2007-086052

  As described above, in the in-vehicle navigation device, even on a special road such as a car pool lane, a guidance route is searched so that the vehicle can pass through an entry permission section between the general lane and the car pool lane. Be able to.

  In Patent Document 1, it is determined whether or not the car pool lane is running, and information on the car pool lane is guided when the car pool lane is running.

  However, there may be a case where it is not possible to accurately determine whether or not the host vehicle is traveling on a car pool lane. For example, even if a specific mark can be detected based on an image taken by an in-vehicle camera, there is a possibility that the mark is marked not on the traveling lane of the host vehicle but on an adjacent lane.

  As described above, it is difficult to determine whether or not the mark is marked on the lane in which the host vehicle is traveling only by detecting the mark.

  Even if it can be detected that a specific mark is present in the traveling lane of the host vehicle, when the host vehicle changes the lane, the mark is detected based on the image acquired after the lane change. It is difficult to determine whether the vehicle is traveling in a specific lane at high speed.

  The present invention has been made in view of the problems of the prior art, and provides a road marking recognition apparatus and a road marking recognition method that can accurately and quickly determine whether a running lane is a specific lane. For the purpose.

  In order to solve the above-described problems of the prior art, according to the basic form of the present invention, a lane mark is detected based on a camera that captures at least a front or rear image of a vehicle and an image captured by the camera. A lane mark detection unit, a specific road marking detection unit that detects a specific road marking based on an image captured by the camera, a travel distance calculation unit that calculates the travel distance of the host vehicle, and a predetermined distance A specific road marking presence determination unit that determines that the host vehicle is traveling in a specific lane when the specific road marking is detected in a lane in which the host vehicle travels, and the lane mark The detection unit identifies a right lane mark and a left lane mark of a lane in which the host vehicle is traveling, and the specific road marking presence determination unit is configured such that a position of the specific road marking is a position of the right lane mark. Position and said When it is between the position of the lane mark on the side, the vehicle is the specific traveling lane and the road sign recognition apparatus, characterized in that the determination is provided.

  In the road marking recognition device according to this aspect, the determination result of the road marking recognition includes a state in which the vehicle is traveling in a specific lane, a state in which the vehicle is traveling in a general lane, and an undefined state in which the traveling lane is not determined. When the specific road marking presence determination unit determines that the vehicle is traveling in a specific lane before traveling the predetermined distance, the travel distance calculation unit calculates the travel distance. You may make it reset, and it may be further provided with a display part, and you may make it display the three states of the determination result of the said road marking recognition.

  The road marking recognition apparatus according to this embodiment further includes a lane change recognition unit that recognizes that the host vehicle has changed the driving lane, and the specific road marking presence determination unit changes the lane to the right lane. The predetermined information in the right lane immediately before the lane change is used as the predetermined information in the traveling lane of the host vehicle after the lane change, and the predetermined information in the traveling lane of the host vehicle immediately before the lane change is changed to the left after the lane change. When the vehicle changes to the left lane, the predetermined information in the left lane immediately before the lane change is used as the predetermined information in the traveling lane of the vehicle after the lane change. The predetermined information in the traveling lane of the host vehicle may be used as the predetermined information in the right lane after the lane change.

  According to another aspect of the present invention, at least a front or rear image of the host vehicle is acquired, and on the basis of the image, the lane mark on the right side of the lane in which the host vehicle is traveling and the left side A step of detecting a lane mark, a step of detecting a specific road marking based on the image, a step of calculating a travel distance of the host vehicle, and a predetermined distance until the vehicle travels a predetermined distance. There is provided a road marking recognition method comprising: a specific road marking presence determination step for determining that the host vehicle is traveling in a specific traveling lane when the road marking is in a predetermined positional relationship. .

  In the road marking recognition method according to this aspect, in the specific road marking presence determination step, when the position of the specific road marking is between the position of the right lane mark and the position of the left lane mark, It may be determined that the vehicle is traveling in the specific lane, and the specific road marking presence determination step determines that the vehicle is traveling in the specific lane before traveling the predetermined distance. The travel distance may be reset.

  In addition, before the specific road marking presence determination step, the vehicle has a lane change recognition step for recognizing that the host vehicle has changed the driving lane, and the specific road marking presence determination step includes changing the lane of the host vehicle to the right lane. The predetermined information in the right lane immediately before the lane change is used as the predetermined information in the traveling lane of the host vehicle after the lane change, and the predetermined information in the traveling lane of the host vehicle immediately before the lane change is changed to the left after the lane change. When the vehicle changes to the left lane, the predetermined information in the left lane immediately before the lane change is set as the predetermined information in the traveling lane of the vehicle after the lane change. There may be included a step of using predetermined information in the travel lane of the host vehicle immediately before the change as predetermined information in the right lane after the lane change.

  According to the road marking recognition apparatus and the road marking recognition method of the present invention, based on the vehicle front or rear image acquired by the in-vehicle camera, the specific mark and the left and right lane marks of the lane in which the host vehicle travels are detected, When it is detected within a predetermined time that the specific mark exists between the left and right lane marks, it is determined that the host vehicle is traveling in the specific lane.

  Thus, since not only the specific mark but also the information of the lane mark is taken into consideration, it is possible to accurately determine whether or not the vehicle is traveling in the specific lane.

  In addition, when the host vehicle changes lanes, the specific mark recognition information and travel distance information in the travel lane of the host vehicle and the left and right lanes before the lane change are changed, and the travel lane and left and right information of the host vehicle after the lane change It is made to become each information in the driving lane.

  As a result, information until the lane change can be used effectively, and it is not necessary to measure the distance traveled again after the lane change, so even if the lane is changed, is the host vehicle traveling in the specified lane? It is possible to determine whether or not at high speed.

It is a block diagram which shows the structure of the vehicle-mounted navigation apparatus which concerns on one Embodiment of this invention. It is a block diagram which shows the structure (the 1) of the road marking recognition apparatus in the vehicle-mounted navigation apparatus of FIG. It is a figure which shows an example of the road which has a car pool lane. It is FIG. (1) explaining the determination process of a car pool lane. It is a flowchart which shows an example of a specific road marking presence determination process. It is a flowchart which shows an example of the detection process of the diamond-type mark in a driving lane. It is a block diagram which shows the structure (the 2) of the road marking recognition apparatus in the vehicle-mounted navigation apparatus of FIG. It is a figure explaining a lane change recognition process. It is FIG. (2) explaining the determination process of a car pool lane. It is a flowchart which shows an example of the specific road marking presence determination process containing a lane change determination process. It is a flowchart which shows an example of a lane change recognition process. It is a flowchart which shows an example of a specific lane driving | running | working recognition process.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

(1) 1st Embodiment (Structure of a vehicle-mounted navigation apparatus)
FIG. 1 is a block diagram showing a configuration of an in-vehicle navigation device 100 according to an embodiment of the present invention.

  In the figure, 1 is a DVD drive, and 1a is a storage medium in which map data and other guidance data are stored. In the present embodiment, the DVD-ROM 1a is used as a storage medium for storing such data, but a hard disk or other storage medium may be used. The map stored here is divided into longitude and latitude widths of appropriate sizes according to each scale level such as 1/1500, 1/25000, 1 / 50,000, 1/10000, etc. Roads, buildings, facilities, and other various properties included in are stored as a coordinate set of points (nodes) expressed in longitude and latitude. Map data includes (1) a road layer composed of road link data, intersection data, node data, etc., (2) a background layer for displaying roads, buildings, parks, rivers, etc. on a map image, and (3) a municipality. It consists of a character / symbol layer for displaying administrative division names such as names, road names, intersection names, characters, map symbols, and the like.

  In the road layer, the road link data supplies attribute information of each road. For each link constituting the road, the type of road (national road, highway, prefectural road, other road), hierarchization of the road network It consists of data such as the level, the number of nodes constituting the link, the road number (road name), and the width of the link connecting each node. The intersection data is a set of nodes closest to the intersection among the nodes on the link coupled to the intersection. The node data is a list of all nodes constituting the road, and includes position information for each node, information for identifying whether the node is an intersection or not, and the like.

  An operation unit 2 is provided with operation buttons and the like for operating the navigation apparatus body 10. In this embodiment, the operation unit 2 includes a remote control transmitter, and the user can also operate the navigation apparatus body 10 with the remote control transmitter at hand.

  Also, 3 receives GPS signals sent from a plurality of GPS satellites and generates and outputs GPS data such as longitude, latitude, PDOP (Position DOP) value and HDOP (Horizontal DOP) value of the current position of the vehicle. A GPS receiver is shown. Reference numeral 4 denotes a self-contained navigation sensor. The self-contained navigation sensor 4 is composed of an angle sensor such as a gyro that detects a vehicle rotation angle, and a travel distance sensor that generates a pulse at every constant travel distance.

  In addition, 5 is a communication device such as an in-vehicle telephone for communicating with various service centers, and 6 is a VICS reception for receiving VICS (registered trademark) (road traffic information communication system) information transmitted from a radio wave beacon or an optical beacon. Indicates the machine. These beacons are installed on the roadside and connected to police stations, road managers and integration centers to provide information on traffic jams in the vicinity.

  Reference numeral 7 denotes a display unit such as a liquid crystal panel. The navigation device body 10 displays a map around the current position of the vehicle on the display unit 7, a guidance route from the departure point to the destination, a vehicle mark, Other guidance information is displayed. The display unit 7 is provided with a touch panel on the screen, and is configured with various buttons paired with the display content of the display screen. The touch panel serves as an input device for selecting a menu or the like indicated by these various buttons. Reference numeral 8 denotes a speaker for providing guidance information to the user by voice.

  Reference numeral 9 denotes an in-vehicle camera disposed at the front center or the rear center of the vehicle, and images a region within a predetermined angle range in front of or behind the vehicle.

  The navigation device body 10 is composed of the following. Reference numeral 11 denotes a buffer memory that temporarily stores map data read from the DVD-ROM 1a via the DVD drive 1.

  Reference numeral 12 denotes a control unit constituted by a microcomputer. The control unit 12 has a built-in navigation program, and calculates the current position of the vehicle based on a signal output from the GPS receiver 3 or a signal output from the self-contained navigation sensor 4 according to the program. Or the map data to be displayed is read from the DVD-ROM 1a to the buffer memory 11 via the DVD drive 1 or from the destination to the destination under the search conditions set using the map data read to the buffer memory 11. Various processes are executed, such as searching for a guide route up to.

  13 is a map drawing unit for generating a map image using the map data read out to the buffer memory 11, and 14 is for generating various menu screens (operation screens) and various marks such as vehicle position marks and cursors according to the operation status. This is the operation screen / mark generator.

  Reference numeral 15 denotes a guide route storage unit that stores the guide route searched by the control unit 12, and 16 denotes a guide route drawing unit that draws the guide route. The guidance route storage unit 15 stores all the nodes of the guidance route searched by the control unit 12 from the departure point to the destination. When displaying the map, the guide route drawing unit 16 reads the guide route information from the guide route storage unit 15 and draws the guide route with a color and line width different from those of other roads.

  Reference numeral 17 denotes a storage unit composed of a hard disk or a semiconductor memory. As will be described later, a determination result as to whether or not a lane in which the host vehicle is traveling is a specific traveling lane, and a diamond type used for the determination. Data such as marks and predetermined characters are stored.

  An audio output unit 18 supplies an audio signal to the speaker 8 based on a signal from the control unit 12. Reference numeral 19 denotes an image composition unit, which overlays the map image drawn by the map drawing unit 13 with various marks and operation screens generated by the operation screen / mark generation unit 14, the guide route drawn by the guide route drawing unit 22, and the like. Display on the display unit 7.

  FIG. 2 is a block diagram showing a configuration of a road marking recognition apparatus 200 that is a part of the in-vehicle navigation apparatus 100 of FIG.

  The road marking recognition apparatus 200 includes a camera 9, a GPS receiver 3, a self-contained navigation sensor 4, a display 7, a storage unit 17, and a control unit 12. The control unit 12 includes a specific road marking recognition unit. A unit 21, a lane mark recognition unit 22, a travel distance calculation unit 23, and a specific road marking presence determination unit 24 are included.

  The specific road marking recognition unit 21 recognizes a specific road marking from the video acquired by the camera 9. The specific road marking is a road marking to be recognized, such as a diamond mark indicating a car pool lane or a character “CAR POOL ONLY”.

  The lane mark recognition unit 22 recognizes the lane mark based on the image acquired by the camera 9. The lane mark is recognized by detecting the straight line by a well-known method, for example, by performing a Hough transform on the image.

  The travel distance calculation unit 23 calculates the travel distance of the vehicle after recognizing a specific road sign.

  The specific road marking presence determination unit 24 determines whether or not the specific road marking recognition unit 21 recognizes the specific road marking in the lane in which the host vehicle recognized by the lane mark recognition unit 22 travels during a predetermined distance. judge.

  In the road sign recognition device 200 configured as described above, an image of a road on which the host vehicle is traveling is acquired by the camera 9, and a specific mark and a lane mark marked on the road are detected based on the image. Thus, it is determined whether or not the host vehicle is traveling in a specific travel lane.

  In addition to detecting specific marks, lane marks are detected, and by detecting the travel position of the host vehicle based on the positional relationship between these marks, it is determined whether the host vehicle is traveling in the specific lane. It becomes possible to determine accurately.

  Hereinafter, a process for determining whether or not the vehicle is traveling in a specific lane for a car pool lane used in the United States as the specific lane will be described with reference to FIGS.

  FIG. 3 is an image in front of the vehicle photographed by the camera 9 and is a diagram showing an example of a road having a car pool lane 33. In FIG. 3, a general lane 34 in which the host vehicle is traveling, a lane mark (white line) 32, and a car pool lane 33 on the left side of the lane mark 32 are displayed.

  The car pool lane 33 is introduced to alleviate traffic congestion, and is a lane that can be used by vehicles on which a plurality of people are riding. In the car pool lane 33, a diamond-shaped mark 31 as shown in FIG. 3 is marked on the road, or the characters “CAR POOL ONLY” are marked. There are several such marks in succession at the entrance to the car pool lane, and after that, they are marked at predetermined intervals, for example, every 1 km.

  4A is an example of an image in front of the vehicle imaged by the camera 9, and FIG. 4B shows whether or not a diamond mark is present in the travel lane based on the image of FIG. 4A. It is a figure explaining the process of. FIG. 4A shows an image in which a diamond mark 43 is marked on a road in a lane in which the host vehicle is traveling, and a left lane mark 41 and a right lane mark 42 are marked.

  FIG. 4B is a diagram showing the distance from the own vehicle to the lane mark that is necessary for determining whether or not the own vehicle is traveling in a specific lane in the image of FIG. is there.

  The image of FIG. 4B is divided into left and right images by an image center line 44. With the image center line 44 as a reference, the right direction of the screen is a positive value, the left direction is a negative value, and the horizontal direction of the image from the image center line 44 to the lane mark (41, 42) (the width direction of the road) The distance (L2, L1) is calculated by image processing.

  Further, a distance L3 from the image center line 44 to the center line 45 of the diamond mark 43 is calculated as the position of the diamond mark 43. When the camera mounting position is deviated from the center of the vehicle, the distance is calculated by correcting the deviation.

  FIG. 4A is an image at a certain point of time taken by the camera 9, and the video taken by the camera 9 is stored in the storage unit 17 as an image every predetermined time, and image processing is performed on each image. Thus, each distance as shown in FIG. 4B is calculated.

  Hereinafter, the specific road marking presence determination process performed by the control unit 12 of the road marking recognition apparatus 200 according to the first embodiment will be described with reference to FIGS. 5 and 6 showing an example of the processing flow and the car pool lane determination process. This will be described with reference to FIG.

  First, initial setting is performed in step S11 of FIG. In the initial setting, various variables used during processing are initialized. The travel distance is calculated using the vehicle speed and time at two time points when the image is acquired (current time and the previous time point when the previous image is acquired). The current vehicle speed (curSpeed) is set to “0”, the vehicle speed at the start of processing is set to the previous vehicle speed (preSpeed), the current time (curTime) is set to “0”, and the previous time (preTime) is processed. Set the start time.

  Further, the travel distance (D) is set to “0”, and a predetermined travel distance used in determining the travel distance is defined as D1. For example, the predetermined traveling distance is an interval (1 km) at which the diamond-shaped mark 43 is marked. It should be noted that the predetermined travel distance D1 may be appropriately set via the operation unit 2 so as to suit the road conditions.

  In addition, “undefined” indicating that a specific lane or a general lane is in an undetermined state is set in the recognition result (result) indicating the result of recognizing the type of travel lane.

  In the next step S <b> 12, the current image captured by the camera 9 and stored in the storage unit 17 is acquired.

  In the next step S13, a lane mark indicated by a solid line or a lane mark indicated by a broken line is detected based on the image acquired in step S12. The lane mark is detected by a well-known method, for example, by detecting a straight line by performing a Hough transform on the image.

  When a lane mark is detected, the distance in the horizontal direction (road width direction) from the lane mark to the host vehicle is calculated. As shown in FIG. 4B, the distances (L1, L2) from the host vehicle to the lane mark are calculated.

  In the next step S14, a specific mark (diamond mark) is detected from the image acquired in step S12. This detection is performed by comparing with a diamond-shaped mark template prepared in advance and stored in the storage unit 17. In addition, although the shape of the diamond-shaped mark acquired according to the distance from a vehicle differs, these different shapes are also prepared as a template.

  When the diamond mark is detected, the distance in the lateral direction (the road width direction) from the center position of the diamond mark to the host vehicle is calculated. As shown in FIG. 4B, a distance L3 from the host vehicle to the diamond mark is calculated.

  In the next step S15, it is determined whether or not there is a diamond mark in the traveling lane of the host vehicle. When this mark is present in the travel lane, the process proceeds to step S21, and when there is no mark, the process proceeds to step S16.

  Here, the diamond type mark detection process in the traveling lane will be described below with reference to the flowchart of FIG.

  In the first step S31, initial setting is performed. In the initial setting, the distance between the left lane and the vehicle (left) is set to “0”, and the distance between the right lane and the vehicle (right) is set to “0”. Further, the index variable (di) of the diamond mark is set to “0”.

  In the next step S32, the maximum negative value among the values indicating the position of the lane mark is set as a variable (left) indicating the left lane.

  In the next step S33, the minimum positive value among the values indicating the position of the lane mark is set as a variable (right) indicating the right lane.

  In the next step S34, it is determined whether or not the index variable (di) is equal to or greater than the number of recognized diamond marks. When the index variable (di) is greater than or equal to the recognized number, the process proceeds to step S37, and when the index variable (di) is less than the recognized number, the process proceeds to step S35.

  In the next step S35, it is determined whether or not the center position of the di-th diamond mark exists in the lane. That is, it is determined whether or not the center position of the di-th diamond mark is larger than the left lane variable left and smaller than the right lane variable right. When it exists in a lane, it transfers to step S36, and when it does not exist in a lane, it transfers to step S38.

  In the next step S36, it is recorded in the storage unit 17 that a diamond mark is present in the traveling lane of the host vehicle.

  On the other hand, when the index variable di is greater than or equal to the number of diamond type marks recognized in step S34, it is recorded in the storage unit 17 in step S37 that there is no diamond type mark in the traveling lane of the host vehicle.

  If it is determined in step S35 that the center position of the di-th diamond mark does not exist in the travel lane of the host vehicle, the index variable di is incremented by 1 in step S38, and the process returns to step S34. continue.

  Returning to FIG. 5, in step S16, the travel distance D of the host vehicle is calculated. The travel distance is calculated based on the speed and time of the host vehicle.

  The current vehicle speed (curSpeed) is set based on information from the GPS receiver 3 and the self-contained navigation sensor 4, and the current time is set as the current time (curTime). Next, the distance from the previous time to the current time is calculated using the vehicle speed (preSpeed) and the time (preTime) at the previous time when the distance was calculated based on these values and the previous image. And added to the variable D indicating the travel distance. That is, D + ((curSpeed + preSpeed) / 2) × (curTime−preTime) is newly set as the travel distance D. In order to calculate the next mileage, the current vehicle speed and the current time are set to the vehicle speed (preSpeed) and the time (preTime) at the previous time, respectively.

  In the next step S17, it is determined whether or not the travel distance D of the host vehicle is equal to or greater than a predetermined travel distance D1. When it is equal to or longer than the predetermined travel distance, the process proceeds to step S18, and when it is less than the predetermined travel distance, the process proceeds to step S19.

  In the next step S18, “general lane” is set in the recognition result variable result, and it is recorded that the lane in which the vehicle is traveling is a general lane.

  On the other hand, when it is determined in step S15 that a diamond mark is present in the travel lane, “carpool lane” is set in the recognition result variable result in step S21, and the lane in which the host vehicle is traveling is determined. Record car lane.

  In the next step S22, the travel distance D of the host vehicle is reset, and the process proceeds to step S19. By resetting the travel distance D to “0”, it is determined that the vehicle has traveled a predetermined distance when the next diamond-shaped mark is detected, even though the vehicle has not traveled a predetermined distance. Identification errors can be prevented.

  In step S19, the recognition result variable result is output as a recognition result. This recognition result may be output to the display unit 7. For example, when it is recognized that the car is driving in the car pool lane, “Currently driving in the car pool lane” is displayed, and when it is recognized that the car is driving in the general lane, "It is driving in a general lane." Is displayed, and when it is uncertain which lane is driving, it is displayed as "Checking the driving lane."

  In the next step S29, it is detected whether or not a navigation end operation has been performed via the operation unit 2. If there is an end operation, the process ends. If there is no end operation, the process returns to step S12 and the process ends. Continue.

  As described above, according to the road marking recognition apparatus and the road marking recognition method of the first embodiment, based on the vehicle front or rear image acquired by the camera 9, the specific mark and the lane in which the host vehicle is traveling When the left and right lane marks are detected and it is detected within a predetermined time that the specific mark exists between the left and right lane marks, it is determined that the host vehicle is traveling in the specific lane.

  Thus, since not only the specific mark but also the information of the lane mark is taken into consideration, it is possible to accurately determine whether or not the vehicle is traveling in the specific lane.

(2) Second Embodiment In the second embodiment, a process for determining at high speed whether or not the vehicle is traveling in a specific lane when the host vehicle changes lanes will be described. In addition, since the structure of the vehicle-mounted navigation apparatus is the same as that of the first embodiment, description thereof is omitted.

  FIG. 7 shows a block diagram of a road marking recognition apparatus 200 in the second embodiment. The difference from the road marking recognition apparatus 200 in the first embodiment is that a lane change recognition unit 25 is added.

  The lane change recognition unit 25 determines whether or not the host vehicle has changed the driving lane based on the captured image. Recognition of the lane change is determined based on a change in the distance from the vehicle to the lane mark in the acquired images at the two time points. The two time points are the current time point and the previous time point. The distance from the current vehicle to the left lane mark is the left lane-to-vehicle distance left, and the current distance from the vehicle to the right lane mark is the right lane-to-vehicle distance. Set to right.

  The distance from the vehicle to the left lane mark at the previous time is set to the left lane-to-vehicle distance preLeft at the previous time, and the distance from the vehicle to the right lane mark at the previous time is set to the right lane-to-vehicle distance preRight at the previous time. Is done.

  The lane change recognition process will be described with reference to FIG. FIG. 8A shows a case where the host vehicle 81 changes the left lane from the lane 82 to the lane 83. In this case, the distance from the own vehicle 81 to the left lane mark 85 before the lane change is set to preLeft, and the distance from the own vehicle 81 to the right lane mark 86 is set to preRight. The distance from the host vehicle 81b to the left lane mark 84 after the lane change is set to left, and the distance from the host vehicle 81b to the right lane mark 85 is set to right.

  If preLeft <priRight and left> right and preLeft <0.4 [m] and right <0.4 [m] at these set values, it is determined that the lane has been changed to the left side. This judgment condition takes into account that the vehicle is close to the lane on the side where the vehicle changes immediately before the lane change, and that the vehicle is close to the lane before the lane change immediately after the lane change. Also, in order to make a more reliable determination, check that the vehicle is on the lane mark on the lane change side immediately before the lane change, and that the vehicle is on the lane mark on the lane before the lane change immediately after the lane change. In consideration of this, the distance between preLeft and right is set.

  FIG. 8B shows a case where the host vehicle 81 changes the right lane from the lane 82 to the lane 84. Also in this case, as in FIG. 8A, the distance from the own vehicle before and after the lane change to the lane mark of each lane is calculated, and it is determined that the right lane has been changed based on the relationship. That is, it is determined that the right lane has been changed when preLeft> preRight and left <right, and preRight <0.4 [m] and left <0.4 [m].

  FIG. 9A is an example of an image in front of the vehicle imaged by the camera, and FIG. 9B shows whether or not a diamond mark is present in the travel lane based on the image of FIG. 9A. It is a figure explaining a process.

  The image in FIG. 9A shows a three-lane road including the own lane and the left and right lanes of the own lane. The diamond type mark (96, 95) is marked on the own lane and the left lane, and the diamond type mark is not marked on the right lane.

  In the specific road marking presence determination process including the lane change process, first, the marked diamond mark and the three-lane lane mark are recognized as in the specific road marking presence determination process of the first embodiment.

  As shown in FIG. 9A, four lane marks 91, 92, 93, 94 for three lanes and two diamond marks 95, 96 are recognized from the acquired image. Based on the position of these recognized marks, the distance from the image center 44 in the horizontal direction (the width direction of the road) is measured with the positive value on the right side and the negative value on the left side. .

  FIG. 9B shows the calculated distance. For each lane mark, a distance L6 from the image center 44 to the lane mark 91, a distance L7 from the image center 44 to the lane mark 92, a distance L8 from the image center 44 to the lane mark 93, and a lane mark 94 from the image center 44. Distance L9 is calculated. For the diamond mark, a distance L4 from the image center 44 to the center line 98 of the diamond mark 95 and a distance L5 from the image center 44 to the center line 97 of the diamond mark 96 are calculated. If the camera mounting position is deviated from the center of the vehicle, the deviation is corrected.

  Hereinafter, the specific road marking presence determination process including the lane change recognition process performed by the control unit 12 will be described with reference to the flowchart of FIG.

  In step S41 of FIG. 10, initial setting is performed. In this initial setting, the recognition result is initialized, the travel distance calculation result is initialized, and the lane change recognition result is initialized. Set the recognition result variable (result [0], result [1], or result [2]) indicating the recognition result for each lane of the left lane, own lane, or right lane to "Indefinite" and initialize it. . Also, set the mileage variable (distance [0], distance [1], or distance [2]) to indicate the mileage in the left lane, own lane, or right lane, respectively. Turn into. In addition, a value “0” indicating that the lane has not been changed is set to the lane change variable cd indicating the recognition result of the lane change and is initialized.

  Further, “0” is set for the left lane-vehicle distance left, the right lane-vehicle distance right, the previous time left lane-vehicle distance preLeft, and the previous time right lane-vehicle distance preRight.

  Also, for each lane, set the current vehicle speed curSpeed to “0”, set the vehicle speed preSpeed to the previous vehicle speed preSpeed, and set the current time curTime to “0” to the previous time pretime. Set the time.

  In the next step S42, the current image captured by the camera 9 and stored in the storage unit 17 is acquired.

  In the next step S43, lane recognition processing is performed based on the image acquired in step S42. Lane recognition is performed by detecting a straight line by a well-known method, for example, by performing a Hough transform on an image.

  In the next step S44, a lane change determination process is performed. As described with reference to FIG. 8, through the lane change variable cd indicating whether or not the own vehicle has changed lanes, and if the lane has been changed, the vehicle has changed to the left lane or the right lane. To obtain the judgment result.

  Here, the lane change recognition process will be described with reference to the flowchart of FIG.

  In step S51 of FIG. 11, initial setting is performed. By default, the absolute value of the distance to the left lane mark on the left side of the current lane is set to the left lane-to-vehicle distance left, and the absolute value of the distance to the right lane mark is set to the right lane-to-vehicle distance right. Set to. Further, the lane change variable cd is set to a value (0) indicating no lane change.

  In the next step S52, it is determined whether or not the lane has been changed to the left lane. As shown in FIG. 8A, the determination is made based on the relationship between the calculated values of preLeft, priRight, left, and right. When the lane is changed to the left lane, the process proceeds to step S54, the value (-1) indicating the lane change to the left lane is set in the lane change variable cd, and the process proceeds to step S56.

  In the next step S53, it is determined whether or not the lane has been changed to the right lane. As shown in FIG. 8A, the determination is made based on the relationship between the calculated values of preLeft, priRight, left, and right. When the lane is changed to the right lane, the process proceeds to step S55, the value (+1) indicating the lane change to the right lane is set in the lane change variable cd, and the process proceeds to step S56.

  In the next step S56, post-processing is performed. That is, the left lane-vehicle distance left is set to the previous time point left lane-vehicle distance preleft, and the right lane-vehicle distance right is set to the previous time point right lane-vehicle distance preright.

  In the next step S57, the value of the lane change variable cd is returned, and this process ends.

  Returning to FIG. 10, in step S <b> 45, it is determined whether or not the host vehicle has changed to the left lane. This determination is performed using the result obtained in step S44. When the host vehicle has changed to the left lane, the process proceeds to step S46, and when the vehicle has not changed to the left lane, the process proceeds to step S47.

  In step S46, a left lane change process is performed. In the left lane change process, the recognition result result [1] of the own lane is the recognition result result [2] of the right lane, the recognition result result [0] of the left lane is the recognition result result [1] of the own lane, and the left lane The recognition result result [0] is undefined because it has not been recognized yet.

  Further, the travel distance distance [1] in the own lane is the travel distance distance [2] in the right lane, the travel distance distance [0] in the left lane is the travel distance distance [1] in the own lane, and the travel distance distance in the left lane. [0] is “0”.

  In the next step S47, it is determined whether or not the host vehicle has changed to the right lane. This determination is performed using the result obtained in step S44. When the own vehicle has changed to the right lane, the process proceeds to step S48, and when the own vehicle has not changed, the process proceeds to step S49.

  In step S48, a right lane change process is performed. In the right lane change process, the recognition result result [1] of the own lane is the recognition result result [0] of the left lane, the recognition result result [2] of the right lane is the recognition result result [1] of the own lane, and the right lane The recognition result result [2] is undefined because it has not been recognized yet.

  Also, the travel distance distance [1] in the own lane is the travel distance distance [0] in the left lane, the travel distance distance [2] in the right lane is the travel distance distance [1] in the own lane, and the travel distance distance in the right lane. [2] is “0”.

  In the next step S49, a specific lane travel recognition process for detecting that the vehicle is traveling in a specific lane is performed. This process will be described later.

  In the next step S50, it is determined whether or not an end operation has been instructed. When there is an end instruction, this process ends. When there is no end instruction, the process returns to step S42 to continue this process.

  Next, the specific lane travel recognition process in step S49 will be described with reference to the flowchart of FIG.

  First, in step S61 in FIG. 12, a specific mark, for example, a diamond mark is recognized. Based on the image acquired in step S42, pattern matching is performed based on a pattern of a diamond-shaped mark prepared in advance, and it is detected whether or not this mark exists in the image.

  In the next step S62, based on the information from the GPS receiver 3 and the autonomous navigation sensor 4, the current vehicle speed is acquired and set to curSpeed, the time is acquired, and the current time curTime is set.

  In the next step S63, the lane number variable LN for identifying the lane is set to “0”. The lane number variable LN is represented by 1 as the traveling lane (own lane) of the host vehicle, 0 as the lane on the left side of the own lane, and 2 as the lane on the right side of the own lane.

  In the next step S64, it is determined whether or not the lane number variable LN is smaller than 3. When it is smaller, the process proceeds to step S65, and when the lane number variable LN becomes 3, the process proceeds to step S82. That is, it is determined whether or not the processing after step S65 has been completed for all of the lane on the left side of the own lane, the own lane, and the right lane of the own lane.

  In the next step S65, it is determined whether or not a specific mark exists in the LNth lane. When the specific mark exists, the process proceeds to step S70, and when the specific mark does not exist, the process proceeds to step S66. The determination of whether or not a specific mark exists is the same as the process of FIG. 6 described in the first embodiment. However, the variable left is set to the distance from the image center 44 to the left lane mark in each lane, and the variable right is set to the distance from the image center 44 to the right lane mark in each lane.

  In the next step S66, the mileage of the LNth lane is calculated. The calculation of the travel distance is the same as the processing in step S16 of FIG. 5 described in the first embodiment.

  In the next step S67, it is determined whether or not the travel distance of the host vehicle is equal to or greater than a predetermined travel distance. When the distance is equal to or longer than the predetermined travel distance, the process proceeds to step S68, and when the predetermined travel distance is not reached, the process proceeds to step S69.

  In the next step S68, the LN-th lane is set to “general lane”.

  On the other hand, in step S70, where the specific mark exists in the LNth lane in step S65, the LNth lane is set to “carpool lane”.

  In the next step S71, the mileage of the LNth lane is reset, and the process proceeds to step S69.

  In the next step S69, the variable LN is incremented by 1 to verify whether or not a specific mark exists for the next lane, and the process proceeds to step S64.

  In step S <b> 72, which shifts when the variable LN becomes 3 in step S <b> 64, the vehicle speed and time are stored as previous data for use in the next process. That is, curTime is set to preTIme and curSpeed is set to preSpeed.

  In the next step S73, the recognition result of the own lane is output via the display unit 7.

  As described above, according to the road marking recognition device and road marking recognition method of the second embodiment, when the host vehicle changes lanes, the driving lane of the host vehicle and the right and left driving lanes before the lane change are changed. The data is shifted so that the recognition information of the specific mark and the information of the travel distance become the respective information in the travel lane of the own vehicle and the left and right travel lanes after the lane change.

  As a result, information until the lane change can be used effectively, and it is not necessary to measure the distance traveled again after the lane change, so even if the lane is changed, is the host vehicle traveling in the specified lane? It is possible to determine whether or not at high speed.

  In the above embodiment, the car pool lane is targeted as the specific lane, and the diamond-shaped mark is the specific road display. However, the present invention is not limited to this, and a specific mark of a predetermined shape or character is at a predetermined interval. When displayed on a road, the present invention is applicable when determining whether or not the host vehicle is traveling on a lane marked with the specific mark.

100: In-vehicle navigation device,
200 ... road marking recognition device,
2 ... operation part,
3 ... GPS receiver,
4 ... Self-contained navigation sensor,
7 ... display part,
9 ... Camera,
10 ... navigation device body,
12 ... control unit,
17 ... storage part,
21 ... Specific road marking recognition unit,
22 ... Lane mark recognition unit,
23: Travel distance calculation unit,
24 ... specific road marking presence determination unit,
25 ... Lane change recognition unit,
31 ... Diamond mark (specific road marking),
33 ... Carpool lane,
34: General lane.

Claims (12)

A camera for taking an image of at least the front or rear of the vehicle;
A lane mark detection unit that detects a lane mark based on an image captured by the camera;
A specific road marking detection unit for detecting a specific road marking based on an image captured by the camera;
A travel distance calculation unit for calculating the travel distance of the host vehicle;
A specific road marking presence determination unit that determines that the host vehicle is traveling in a specific lane when the specific road marking is detected in a lane in which the host vehicle is traveling before traveling a predetermined distance;
With
The lane mark detection unit identifies a lane mark on the right side and a lane mark on the left side of the lane in which the host vehicle is traveling,
The specific road marking presence determination unit, when the position of the specific road marking is between the position of the right lane mark and the position of the left lane mark, the host vehicle travels in the specific lane. A road marking recognition device characterized by determining that the vehicle is present.   The determination result of the road marking recognition is one of a state in which the vehicle is traveling in a specific lane, a state in which the vehicle is traveling in a general lane, and an indeterminate state in which the traveling lane is not determined. Item 4. A road marking recognition apparatus according to Item 1.   3. The travel distance calculation unit resets the travel distance when the specific road marking presence determination unit determines that the vehicle travels in a specific lane before traveling the predetermined distance. The road marking recognition device described in 1. In addition, the display unit
The road marking recognition apparatus according to claim 3, wherein three states of the determination result of the road marking recognition are displayed. Furthermore, an operation unit for inputting user instructions is provided.
The road sign recognition apparatus according to claim 1, wherein the predetermined distance is set via the operation unit. Furthermore, the vehicle has a lane change recognition unit that recognizes that the host vehicle has changed the driving lane,
The specific road marking presence determination unit, when the own vehicle changes to the right lane, changes the predetermined information in the right lane immediately before the lane change to the predetermined information in the traveling lane of the own vehicle after the lane change, and changes the lane Predetermined information in the driving lane of the vehicle immediately before is set as predetermined information in the left lane after the lane change,
When the host vehicle changes to the left lane, the predetermined information in the left lane immediately before the lane change is used as the predetermined information in the driving lane of the host vehicle after the lane change, and the predetermined information in the driving lane of the host vehicle immediately before the lane change is changed. The road marking recognition apparatus according to claim 1, wherein the information is used as predetermined information in the right lane after the lane change.   The road marking recognition apparatus according to claim 6, wherein the predetermined information includes a recognition result of a specific marking and a travel calculation distance number. Acquiring at least a front or rear image of the vehicle;
Detecting a lane mark on the right side and a lane mark on the left side of the lane in which the host vehicle is traveling based on the image;
Detecting a specific road sign based on the image;
Calculating the mileage of the host vehicle;
A specific road marking presence determination step for determining that the host vehicle is traveling in a specific traveling lane when the lane mark and the specific road marking are in a predetermined positional relationship before traveling a predetermined distance;
A road marking recognition method characterized by comprising:   In the specific road marking presence determination step, when the position of the specific road marking is between the position of the right lane mark and the position of the left lane mark, the host vehicle travels in the specific lane. The road marking recognition method according to claim 8, wherein the road marking recognition method is determined.   9. The road marking according to claim 8, wherein, in the specific road marking presence determination step, when it is determined that the vehicle is traveling in a specific lane before traveling the predetermined distance, the travel distance is reset. Recognition method. Before the specific road marking presence determination step, the vehicle has a lane change recognition step for recognizing that the host vehicle has changed the driving lane,
The specific road marking presence determination step includes
When the host vehicle changes to the right lane, the predetermined information in the right lane immediately before the lane change is used as the predetermined information in the driving lane of the host vehicle after the lane change, and the predetermined information in the driving lane of the host vehicle immediately before the lane change is set. Making the information of the predetermined information in the left lane after the lane change,
When the host vehicle changes to the left lane, the predetermined information in the left lane immediately before the lane change is used as the predetermined information in the driving lane of the host vehicle after the lane change, and the predetermined information in the driving lane of the host vehicle immediately before the lane change is changed. The road marking recognition method according to claim 8, further comprising a step of making the information of the predetermined information in the right lane after the lane change.   The road marking recognition method according to claim 11, wherein the predetermined information includes a recognition result of a specific marking and a calculated number of travel distances.

Images