JP2007011937A - Signal detection system, signal detection device, information center, and signal detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To identify the location of a signal at an intersection. <P>SOLUTION: A navigation apparatus 100 transmits the location of a signal detected with an imaging apparatus to an information center 200. The information center 200 statistically processes the received location of the signal to identify the location of the signal, and distributes the location of the identified signal to the navigation apparatus 100. In addition, the navigation apparatus has a means for detecting a signal for pedestrians, and notifies the status of the signal to a driver, according to the status of the signal for pedestrians. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a traffic signal detection apparatus, a traffic signal detection method, and a traffic signal detection system for detecting the position of a traffic signal at an intersection.

  The following stop line detection device for a vehicle is known from Patent Document 1. According to this vehicle stop line detection device, the stop line position at the intersection is detected based on the imaging device that images the front of the vehicle and the road recapture data or the stop line width data stored in the map data.

JP 2004-295597 A

  However, when the conventional method for detecting a stop line is applied to the detection of traffic lights at intersections, it is necessary to use information stored in map data. In addition, there is a problem that the detection accuracy of the traffic light may be lowered when the map data is old.

  The present invention detects a traffic signal included in an image captured by the imaging means, calculates the position of the detected traffic signal on the map, transmits the calculated traffic signal position on the map to the information center, The received traffic signal on the map is statistically processed to identify the traffic signal on the map, and the identified traffic signal on the map is distributed.

  According to the present invention, the calculated position of the traffic light on the map is transmitted to the information center, and the information center statistically processes the position of the received traffic light on the map to identify the position of the traffic light on the map. The position of the traffic light on the map was distributed. As a result, it is not necessary to store information about all intersections in advance in the map data, and it is possible to always specify the latest intersection position.

-First embodiment-
FIG. 1 is a block diagram showing a configuration of an embodiment of a traffic light detection system in the first embodiment. In the traffic signal detection system 1, a navigation device 100 as a traffic signal detection device and an information center 200 that distributes traffic signal position information to the navigation device are connected by a predetermined communication line.

  The navigation device 100 includes a vehicle speed sensor 101 that detects the vehicle speed of the host vehicle, a GPS unit 102 that receives a GPS signal from a GPS (Global Positioning System) satellite and detects the current position of the host vehicle, and a traveling direction of the host vehicle. A gyro sensor 103 for detecting the steering angle, a steering angle sensor 104 for detecting the steering angle of the steering wheel, a tilt sensor 105 for detecting the tilt angle of the host vehicle, and a front camera 106 installed so as to image the front of the host vehicle. , A rear camera 107 installed to image the rear of the host vehicle, a control device 108 composed of a memory, a CPU, and other peripheral circuits, a monitor 109 for displaying map information and various operation menus, and a voice Speaker 110 for output and map data storage memory 111 for storing map data And a communication device 112 for communicating with an information center 200 to be described later.

  The map data storage memory 111 stores map data of a plurality of scales (scale ratios), and the stored map data includes road data representing roads by nodes and links. In this road data, A node corresponds to the intersection, and a line segment connecting each node corresponds to a link. In this embodiment, a node ID is assigned to each node as an identification number, and a link ID (road number) is assigned to each link as an identification number. An arbitrary point on the map data can be specified by latitude and longitude information on the map data.

  In addition to road data, the map data also includes information regarding the installation position of traffic lights and the installation positions of traffic lights for pedestrians as intersection information corresponding to each node. And the information regarding the installation position of this traffic light is updated at any time by the latest traffic light position information distributed from the information center 200 mentioned later, and the newest position information is always stored.

  The control device 108 includes a traffic light detection unit 108a, a traffic light position calculation unit 108b, an information transmission unit 108c, and a guidance output unit 108d.

  The traffic signal detection unit 108a detects the approach of the host vehicle to an intersection where there is a traffic signal based on the current position of the host vehicle output by the GPS unit 102 and the intersection information included in the map data. When it is detected that the host vehicle enters an intersection with a traffic signal, the front camera 106 is controlled to capture an image (image 1) in front of the host vehicle, and is taken into a memory area of the control device 108. Then, image analysis is performed on the image 1 to detect a traffic light included in the image 1.

  In detecting a traffic light included in the image 1, for example, the image 1 is subjected to a matching process with a traffic light shape pattern stored in advance, and a shape matching the traffic light shape pattern is extracted from the image 1. Thereby, for example, as shown in FIG. 2A, the traffic lights 2a and 2b existing in front of the host vehicle can be detected.

  The signal detection unit 108a further controls the front camera 106 when the host vehicle travels a predetermined distance after capturing the image 1 described above and turns right or left at the intersection, and is present at the left or right turn. An image including a traffic signal is captured. In addition, the predetermined distance which the own vehicle moved after imaging the image 1 as a timing which images the image including the traffic light which exists in the left turn or the right turn is assumed to be set in advance. Then, based on outputs from the vehicle speed sensor 101, the gyro sensor 103, the rudder angle sensor 104, and the tilt sensor 105, it is determined whether or not the host vehicle has traveled this predetermined distance in the left turn direction or the right turn direction.

  For example, as shown in FIG. 2B, when the host vehicle makes a left turn at an intersection, an image 2 including a traffic light 2c existing ahead of the left turn destination can be captured. The same image analysis as that of the image 1 is performed, and the traffic light 2c included in the image 2 is detected.

  Thereafter, when the host vehicle further travels a predetermined distance and moves to a position where the rear camera 107 can capture an image including a rear traffic signal, the rear camera 107 is controlled to acquire an image (image 3) behind the host vehicle. To do. The travel distance of the host vehicle after capturing the image 2 as the image capture timing at this time is also set in advance, and the actual travel distance of the host vehicle is the vehicle speed sensor 101, the gyro sensor 103, the steering angle sensor 104, and the tilt angle. This is calculated based on the output from the sensor 105.

  As a result, for example, as shown in FIG. 2 (c), an image 3 including traffic lights 2 d and 2 e existing behind the host vehicle after turning left at the intersection can be captured. By performing image analysis, the traffic lights 2d and 2e included in the image 3 can be detected.

  When a traffic light is detected by image analysis from within each of the images 1 to 3 described above, the traffic light position calculation unit 108b calculates a relative position of each traffic light with respect to the vehicle position at the time of image capturing. That is, the direction of each traffic signal viewed from the host vehicle and the distance to each traffic signal are calculated. Specifically, the direction from the own vehicle of the traffic signal detected in each image is calculated based on camera parameters such as the viewing axis direction and the angle of view of the front camera 106 and the rear camera 107 that are measured in advance. Further, the size of the traffic light imaged from a point separated by a certain distance, for example, the horizontal width is compared with the traffic light size detected from within each image, and the distance from the host vehicle to the traffic light is calculated based on the size ratio.

  Then, based on the position information of the own vehicle detected by the GPS unit 102 at the time of capturing each image and the relative position from the own vehicle of the traffic signal included in each image calculated by the above-described processing, Identify the latitude and longitude information of the traffic light. Thereby, the installation positions of a plurality of traffic lights installed in each direction of the intersection where the host vehicle has traveled can be specified.

  The information transmission unit 108c transmits information regarding the traffic signal installation position at the intersection calculated by the traffic signal position calculation unit 108b to the information center 200 via the communication device 112 as intersection information.

  The guidance output unit 108d is a speaker when the traffic signal detection unit 108a detects that the host vehicle enters the intersection where the traffic signal is installed based on the traffic signal position information stored in the map data storage memory 111. Via 110, the guidance which notifies a driver | operator that it will approach to the intersection where the traffic signal is installed is output. Thus, the driver can know in advance that there is an intersection with a traffic signal at the destination.

  The information center 200 includes a control device 201 including a memory, a CPU, and other peripheral circuits, a communication device 202 for communicating with the navigation device 100, and an intersection information storage memory that stores intersection information received from the navigation device 100. 203 and traffic signal position information for storing accurate traffic signal position information calculated by a traffic signal position information calculation unit 201a, which will be described later, as traffic signal position information based on intersection information stored in the intersection information storage memory 203 A storage memory 204 and a map database 205 for storing the map data described above are provided.

  The control device 201 includes a traffic signal position information calculation unit 201a and an information distribution unit 201b. The traffic signal position information calculation unit 201 a statistically processes the intersection information of each intersection stored in the intersection information storage memory 203 to calculate accurate traffic signal position information. That is, when the intersection information of the same intersection is received from the navigation device 100 mounted on a plurality of vehicles, an error may occur in the detection result of the traffic signal installation position by each navigation device 100. For example, an error may occur in the calculated traffic signal position due to a difference in the passing speed of each vehicle or a difference in weather when traveling at the intersection.

  In order to eliminate such an error, accurate position information on the map of the traffic signal is calculated using a general statistical processing technique. For example, accurate position information is calculated by calculating an average value of latitude and longitude information of each traffic signal detected by a plurality of navigation devices 100. Then, accurate traffic signal position information at the intersection is stored in the traffic signal position information storage memory 204 as traffic signal position information.

  When the latest traffic signal position information is calculated by the traffic signal position information calculation unit 201a and stored in the traffic signal position information storage memory 204, the information distribution unit 201b updates the intersection information of the map data stored in the map database 205. At the same time, it is delivered to the navigation device 100 via the communication device 202. Thereby, the intersection information in the map data storage memory 111 of the navigation device 100 can also be updated with the latest traffic signal position information, and the guidance output unit 108d alerts the driver based on the latest traffic signal position information. It can be performed.

  FIG. 3 is a flowchart showing processing of the navigation device 100 according to the first embodiment. The processing shown in FIG. 3 is executed by the control device 108 as a program that starts when the navigation device 100 is powered on.

  In step S <b> 10, the traffic signal detection unit 108 a determines the traffic signal of the host vehicle based on the current position of the host vehicle output by the GPS unit 102 and the intersection information included in the map data stored in the map data storage memory 111. Detecting entry to an intersection with Then, it progresses to step S20 and it is judged whether the own vehicle approachs the intersection with a traffic light. If it is determined that the host vehicle enters an intersection with a traffic light, the process proceeds to step S30.

  In step S30, the guidance output unit 108d outputs a guidance for notifying the driver that the vehicle will enter the intersection where the traffic light is installed, via the speaker 110. Thereafter, the process proceeds to step S40, and the traffic light detection unit 108a controls the front camera 106 to capture an image ahead of the host vehicle (image 1), captures it in the memory area of the control device 108, and proceeds to step S50. The image analysis described above is performed on the image 1 to detect a traffic signal included in the image 1. Thereafter, the process proceeds to step S60.

  In step S60, as described above, it is determined whether or not the vehicle has traveled a predetermined distance after capturing the image 1. If it is determined that the vehicle has traveled the predetermined distance, the process proceeds to step S70. In step S70, the front camera 106 is controlled to capture an image (image 2) in front of the host vehicle, the image is taken into the memory area of the control device 108, and the process proceeds to step S80. To detect traffic lights included in the image 2. Thereafter, the process proceeds to step S90.

  In step S90, as described above, it is determined whether or not the vehicle has traveled a predetermined distance after capturing the image 2. If it is determined that the vehicle has traveled the predetermined distance, the process proceeds to step S100. In step S100, the rear camera 107 is controlled to take an image (image 3) behind the host vehicle, and the image is taken into the memory area of the control device 108. The process proceeds to step S110, and the above-described image analysis is performed on this image 3. To detect traffic lights included in the image 3. Then, it progresses to step S120.

  In step S120, the direction and distance from the own vehicle position at the time of capturing each of the images 1 to 3 to each traffic light detected in each image is calculated, and the installation position of each traffic light is specified. Then, it progresses to step S130 and the information transmission part 108c transmits the information regarding the installation position of the traffic signal in the calculated intersection to the information center 200 via the communication apparatus 112 as intersection information. Thereafter, the process proceeds to step S140.

  In step S140, it is determined whether the traffic signal position information described above has been received from the information center 200. If it is determined that the traffic signal position information has been received, the process proceeds to step S150, and the intersection information of the map data stored in the map data storage memory 111 is updated based on the received traffic signal position information. Then, it progresses to step S160, it is judged whether the power supply of the navigation apparatus 100 was turned off, and when it is judged that it is not turned off, it returns to step S20 and repeats a process. On the other hand, if it is determined that it has been turned off, the process ends.

  FIG. 4 is a flowchart showing the processing of the information center 200 in the first embodiment. The process shown in FIG. 4 is executed by the control device 201 as a program that starts when the information center 200 is powered on.

  In step S <b> 210, the traffic light position information calculation unit 201 a determines whether the intersection information described above has been received from the navigation device 100. If it is determined that the intersection information has been received, the process proceeds to step S220. In step S220, as described above, statistical processing is performed based on the latitude / longitude information of each traffic light detected by the plurality of navigation devices 100, and the accurate position information of each traffic light is calculated, and the traffic signal position information is obtained as traffic light position information. Store in the information storage memory 204. Thereafter, the process proceeds to step S230.

  In step S230, the information distribution unit 201b updates the intersection information of the map data stored in the map database 205 based on the calculated traffic signal position information, and proceeds to step S240. In step S240, the calculated traffic signal position information is distributed to the navigation device 100 via the communication device 202. Thereafter, the process proceeds to step S250.

  In step S250, it is determined whether the information center 200 is turned off. If it is determined that the power is not turned off, the process returns to step S210 to repeat the process. On the other hand, if it is determined that the power is turned off, the process is terminated.

According to the first embodiment described above, the following operational effects can be obtained.
(1) By detecting the traffic light from the captured image and calculating the position of the traffic signal with respect to the vehicle position at the time of image capture, the latitude / longitude information of the traffic light is calculated and the installation position of the traffic light within the intersection is specified. I made it. As a result, it is possible to detect the traffic light from the image actually taken by the vehicle and to identify the position thereof, and to always grasp the installation position of the latest traffic light.

(2) The intersection information detected by the navigation device 100 mounted on each vehicle is transmitted to the information center 200, and the information center 200 statistically processes the intersection information received from the plurality of navigation devices 100 to obtain an accurate traffic signal. The installation position of was calculated. Accordingly, it is possible to correct the traffic signal position error caused by the difference in the passing speed of each vehicle and the weather at the time of traveling at the intersection, and to calculate the accurate traffic signal position information.

(3) The traffic signal position information calculated in the information center 200 is distributed to the navigation device 100, and the intersection information of the map data stored in the map data storage memory 111 is updated based on the received traffic signal position information. I made it. Thereby, the intersection information of the map data stored in the map data storage memory 111 can always be kept up-to-date.

(4) When it is detected that the host vehicle enters an intersection where a traffic signal is installed, a guidance is sent to notify the driver that the vehicle will enter an intersection where a traffic signal is installed. I tried to do it. Thus, the driver can know in advance that there is an intersection with a traffic signal at the destination.

(5) The traffic signal detection unit 108a performs the above-described various processes when it detects that the host vehicle enters the intersection where the traffic signal is installed. Thus, it is only necessary to perform processing on the intersection where the traffic light is installed, and execution of unnecessary processing can be prevented.

-Second embodiment-
In the navigation device according to the second embodiment, in addition to the signal detection processing described above in the first embodiment, when the own vehicle enters an intersection where a pedestrian signal is installed, the navigation device is installed in front of the own vehicle. Pedestrian signal being detected. Furthermore, the state of the detected pedestrian signal light, that is, whether the signal is blue, red, or blinking blue is determined and notified to the driver. Note that the processing of the information center 200 shown in FIG. 4 is the same as that of the first embodiment, and thus the description thereof is omitted.

  FIG. 5 is a block diagram showing a configuration of an embodiment of a traffic light detection system according to the second embodiment. In FIG. 5, the same components as those in the first embodiment shown in FIG.

  The control device 108 of the navigation device 100 includes a pedestrian traffic light state determination unit 108e that determines the state of the pedestrian signal.

  When it is determined that the host vehicle enters the intersection where the traffic signal is installed, the signal detection unit 108a is based on the current position of the host vehicle output by the GPS unit 102 and the intersection information included in the map data. It is determined whether a pedestrian traffic light is installed at the intersection. When it is determined that the pedestrian traffic light is installed, the pedestrian traffic signal is detected from the image 1 by the image analysis described above. For example, as shown in FIG. 6, a pedestrian traffic light 6a installed in front of the vehicle at the intersection is detected.

  The pedestrian traffic light state determination unit 108e determines the state of the pedestrian traffic light 6a by identifying the color of the signal light in the pedestrian traffic light 6a detected by the traffic signal detection unit 108a. In order to identify blue blinking of the signal lamp, a plurality of images are taken at a minute predetermined time interval, and the color change state is also determined.

  When the state of the pedestrian traffic light 6a in front of the host vehicle determined by the pedestrian traffic signal state determination unit 108e is blinking blue, the guidance output unit 108d immediately changes the traffic signal ahead of the host vehicle to red. The driver will be notified accordingly. As a result, the driver can recognize in advance that the signal of the destination of the vehicle will change to red in a short time, and can take actions to avoid danger, such as decelerating in advance. become able to.

  As in the first embodiment, the traffic signal position calculation unit 108b calculates the direction and distance to the pedestrian traffic signal 6a with respect to the vehicle position, and obtains the latitude and longitude information of the pedestrian traffic signal 6a on the map data. Identify. And the information transmission part 108c includes the latitude longitude information of this pedestrian traffic light 6a in the intersection information mentioned above, and transmits to the information center 200. When the information center 200 receives intersection information including latitude / longitude information of the same pedestrian traffic light from a plurality of navigation devices 100, the information center 200 performs statistical processing based on the latitude / longitude information to obtain accurate pedestrian information. Locate the traffic light.

  FIG. 7 is a flowchart showing the processing of the navigation device 100 according to the second embodiment. The processing shown in FIG. 3 is executed by the control device 108 as a program that starts when the navigation device 100 is powered on. In FIG. 7, the same processing numbers as those in the first embodiment shown in FIG. 3 are given the same step numbers, and differences will be mainly described.

  In step S51, a pedestrian traffic light determination process described later in FIG. 8 is executed. In the pedestrian traffic light determination process shown in FIG. 8, in step S310, the traffic light detection unit 108a is based on the current position of the own vehicle output by the GPS unit 102 and the intersection information included in the map data. It is determined whether or not a pedestrian traffic light is installed at the intersection where the car enters. If it is determined that no pedestrian traffic light is installed, the process returns to the process shown in FIG. In contrast, if it is determined that a pedestrian traffic light is installed, the process proceeds to step S320.

  In step S320, as described above, image analysis is performed on the image 1 captured by the front camera 106, and a pedestrian traffic light is detected from the image 1. Then, it progresses to step S330, the pedestrian traffic light state determination part 108e determines the state of the signal light of the detected pedestrian traffic light, and progresses to step S340. In step S340, it is determined whether or not the state of the signal lamp is blinking blue. When it is determined that the state of the signal lamp is not blinking blue, the process returns to the process shown in FIG. On the other hand, if it is determined that the state of the signal lamp is blinking blue, the process proceeds to step S350.

  In step S350, the guidance output unit 108d outputs a guidance for notifying that the pedestrian traffic light is blinking in blue, and returns to the process shown in FIG.

  According to the second embodiment described above, the following effects can be obtained in addition to the functions and effects of the first embodiment. That is, the state of the signal light of the traffic light for pedestrians existing in front of the host vehicle is determined, and when the state of the signal light is blinking blue, guidance for notifying that is output. As a result, the driver can recognize in advance that the signal of the destination of the vehicle will change to red in a short time, and can take actions to avoid danger, such as decelerating in advance. become able to.

-Modification-
In addition, the navigation apparatus of embodiment mentioned above can also be deform | transformed as follows.
(1) In the embodiment described above, the intersection information calculated by the navigation device 100 is transmitted to the information center 200 via the communication device 112, and the information center 200 calculates accurate traffic signal location information, that is, traffic signal location information. An example of distribution to the navigation device 100 has been described. However, the present invention is not limited to this, and the navigation apparatus 100 having no communication function may accumulate intersection information based on the traveling result of the host vehicle, and calculate traffic signal position information based on the accumulated intersection information. That is, the navigation device 100 alone may have all the functions of the traffic light detection system 1 described above.

(2) In the above-described embodiment, when detecting a traffic light included in an image, for example, matching processing with a traffic light shape pattern stored in advance is performed on the image, and the traffic light shape pattern is detected from within the image. An example of extracting matching shapes has been described. However, the present invention is not limited to this, and a traffic signal included in the image may be detected by other object detection methods.

(3) In the above-described embodiment, the case where the traffic signal detection device is, for example, a navigation device has been described. However, the present invention is not limited to this, and other information devices including a function for detecting the vehicle position and map data are provided. You may apply to.

  Note that the present invention is not limited to the configurations in the above-described embodiments as long as the characteristic functions of the present invention are not impaired.

  The correspondence between the constituent elements of the claims and the embodiment will be described. The traffic light detection unit 108a corresponds to a pedestrian traffic light detection unit, and the traffic signal position information calculation unit 201a corresponds to a position specifying unit. The above description is merely an example, and when interpreting the invention, there is no limitation or restriction on the correspondence between the items described in the above embodiment and the items described in the claims.

It is a block diagram which shows the structure of one Embodiment of the traffic signal detection apparatus in 1st Embodiment. It is a figure which shows the example of a traffic light imaging by the front camera 106 and the back camera 107. FIG. It is a flowchart figure which shows the process of the navigation apparatus 100 in 1st Embodiment. It is a flowchart figure which shows the process of the information center 200 in 1st Embodiment. It is a block diagram which shows the structure of one Embodiment of the traffic signal detection apparatus in 2nd Embodiment. It is a figure which shows the example of an imaging of the pedestrian traffic light by the front camera. It is a flowchart figure which shows the process of the navigation apparatus 100 in 2nd Embodiment. It is a flowchart figure which shows the traffic signal determination process for pedestrians in 2nd Embodiment.

Explanation of symbols

1 Signal detection system 100 Navigation device (signal detection device)
DESCRIPTION OF SYMBOLS 101 Vehicle speed sensor 102 GPS unit 103 Gyro sensor 104 Steering angle sensor 105 Tilt sensor 106 Front camera 107 Rear camera 108, 201 Control device 108a Traffic light detection part 108b Traffic light position calculation part 108c Information transmission part 108d Guidance output part 108e Pedestrian traffic light state Determination unit 109 Monitor 110 Speaker 111 Map data storage memory 112, 202 Communication device 200 Information center 201a Traffic signal position information calculation unit 201b Information distribution unit 203 Intersection information storage memory 204 Traffic signal position information storage memory 205 Map database

Claims (6)

A traffic signal detection system in which a traffic signal detection device mounted on a plurality of vehicles and an information center for specifying a position of a traffic signal detected by each traffic signal detection device on a map are connected by a communication line,
Each of the signal detectors is
A traffic light detection means for detecting a traffic light included in an image captured by the imaging means;
Position calculating means for calculating the position on the map of the traffic light detected by the traffic light detecting means;
Transmission means for transmitting the position on the map of the traffic light calculated by the position calculation means to the information center,
The information center
Statistically processing the positions of the plurality of traffic lights received from the respective traffic light detection means, and specifying the positions of the traffic lights on the map; and
A traffic signal detection system comprising: distribution means for distributing a traffic signal position specified by the position specification means to the traffic signal detection device. The traffic light detection system according to claim 1,
The traffic light detection means includes
Based on the map data, including intersection detection means for detecting the entry of the vehicle to the intersection where the traffic light is installed,
A traffic signal detection system for detecting a traffic signal included in an image captured by the imaging unit when the intersection detection unit detects that the host vehicle enters an intersection where a traffic signal is installed. In the traffic light detection system according to claim 1 or 2,
Each of the signal detectors is
Based on the map data, further comprising a pedestrian traffic light detection means for detecting a pedestrian traffic light installed at the intersection,
When the pedestrian traffic light is detected by the pedestrian traffic light detection means, signal light state determination means for determining the state of the signal light of the pedestrian traffic light;
A traffic light detection system further comprising notification means for notifying a driver of a signal change based on a determination result by the signal light state determination means.   The traffic signal detection apparatus as described in any one of Claims 1-3.   The information center according to any one of claims 1 to 3. Detect traffic lights included in the image captured by the imaging means,
Calculate the position of the detected traffic light on the map,
Send the calculated traffic signal position on the map to the information center,
The information center statistically processes the position of the received traffic signal on the map to identify the position of the traffic light on the map,
A traffic signal detection method, comprising: distributing a position of a specified traffic signal on a map.

Images