JP3953858B2 - Car navigation system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an in-vehicle navigation device that guides a vehicle to a destination, and more particularly to an in-vehicle navigation device having an image recognition device.
[0002]
[Prior art]
The in-vehicle navigation device is a map data storage device such as a DVD (Digital Versatile Disk) on which map data is recorded, a display device, a gyro, a GPS (Global Positioning System) receiver, a vehicle speed sensor and other current positions and It includes a vehicle movement detection device that detects the direction. Then, map data including the current position of the vehicle is read from the map data storage device, a map image around the vehicle position is drawn based on the map data, and a vehicle position mark (location) is displayed superimposed on the map image. By scrolling the map image as the vehicle moves, or fixing the map image to the screen and moving the vehicle position mark, you can see at a glance where the vehicle is currently traveling Yes.
[0003]
In general, the in-vehicle navigation device is equipped with a route guidance function that allows the user to easily travel to a desired destination without making a mistake on the road. According to this route guidance function, the lowest cost route connecting from the starting point to the destination using map data is automatically searched by performing simulation calculation such as horizontal search method or Dijkstra method, and the searched route is guided. Memorize it as a route and draw a guidance route on the map image with a different color from other roads and display it on the screen while driving, or a fixed distance to the intersection where the vehicle should change the route on the guidance route When the user approaches the inside, the user is guided to the destination by drawing an arrow indicating the route at the intersection where the route on the map image should be changed and displaying it on the screen.
[0004]
Cost is based on distance, road width, road type (general road or highway, etc.), a value multiplied by a constant according to right turn and left turn, etc., estimated travel time of vehicle, etc. The degree of appropriateness as a numerical value. Even if there are two routes having the same distance, the cost differs depending on whether the user uses a toll road or whether to give priority to distance or time.
[0005]
A map stored in a map data storage device such as a DVD is divided into longitude and latitude widths of appropriate sizes according to the scale level such as 1/12500, 1/25000, 1/50000, and 1/100000. A road or the like is stored as a coordinate set of vertices (nodes) expressed by longitude and latitude. A road is formed by connecting two or more nodes, and a portion connecting two nodes is called a link. The map data includes (1) a road layer including a road list, a node table, an intersection configuration node list, and the like. (2) a background layer for displaying roads, buildings, parks, rivers, and the like on the map screen. 3) Consists of character / symbol layers for displaying characters such as administrative division names such as municipalities, road names, intersection names, building names, and map symbols.
[0006]
[Problems to be solved by the invention]
In-vehicle navigation devices use techniques such as GPS, self-contained navigation, map matching, and the like to improve vehicle position detection accuracy. However, if the highway and the general road are parallel, or if there is a highway above or below the general road, the navigation device is traveling on the highway even though it is traveling on the general road. In some cases, the navigation device may determine that the vehicle is traveling on a general road even though the vehicle is traveling on a highway. Since the guidance information provided by the navigation device differs depending on whether the vehicle is traveling on a general road or on a highway, if the navigation device misidentifies the traveling road, incorrect guidance information is provided to the driver. The
[0007]
Accordingly, an object of the present invention is to provide a vehicle-mounted navigation device that can more accurately determine the type of road on which a vehicle is traveling.
[0008]
[Means for Solving the Problems]
An in-vehicle navigation device according to the present invention includes a map data storage unit that stores map data, a vehicle position detection unit that detects a current position of the vehicle, and a scenery outside the vehicle in the in-vehicle navigation device that guides the vehicle to a destination. And image recognition processing of the image, and for each image, an image recognition device that checks the presence / absence of an object specific to the expressway and the presence / absence of an object specific to a general road from the vehicle position detection means to the vehicle Current position data is input, the map data stored in the map data storage means is used to provide guidance information to the vehicle driver, and the road of the road on which the vehicle is traveling is referenced with reference to the output of the image recognition device. The image recognition apparatus has a database in which priority is set for each object unique to the expressway and each object unique to the general road. Then, a point corresponding to the priority is given to the object detected by the image recognition process, and the reliability of the determination result of the road type is evaluated.
[0009]
In the present invention, a scene outside the vehicle is photographed, and the image is subjected to image recognition processing by an image recognition device, so that objects unique to the expressway (for example, signs such as service areas, side walls, roadside white lines, etc.) The presence / absence and the presence / absence of an object specific to a general road (for example, a sign such as a stop and a crosswalk) are checked. When an object specific to the highway is detected by the image recognition device and no object specific to the general road is detected, it can be said that the vehicle is traveling on the highway. Conversely, when an object specific to a general road is detected by the image recognition apparatus and no target specific to a highway is detected, it can be said that the vehicle is traveling on the general road.
[0010]
However, since it is difficult to determine the type of road on which the vehicle is traveling with one image, it is preferable to determine the road type from the determination results for a plurality of images. For example, every time the vehicle travels a certain distance, a scene outside the vehicle is photographed, and the presence or absence of each object may be determined from a plurality of consecutive images.
[0011]
In this case, the effective appearance rate of each object is recorded in the image recognition device, the appearance rate of each object is calculated from the determination results of a plurality of images, and the calculation result is compared with the effective appearance rate. Thus, it is preferable to determine the presence or absence of each object. Thereby, the detection accuracy of a target object improves.
[0012]
In addition, it is possible to set the priority for each object, assign points according to the priority to the object detected by the image recognition process, and evaluate the reliability of the road type determination result. For example, when a highway side wall or roadside white line is detected, the road being traveled is likely to be an expressway, and when a traffic light or a pedestrian crossing is detected, the road being traveled may be a general road Therefore, when these objects are detected, data indicating that the reliability of the determination result is high is added to the determination result data. As a result, the control means preferentially considers the determination result by the image recognition apparatus and determines the road type of the running road.
[0013]
In the present invention, the control means considers the output of the image recognition device only when there is a possibility that the highway and the general road are mistaken, that is, when the highway and the general road are traveling in an adjacent area. It is preferable to determine the type.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
[0015]
FIG. 1 is a block diagram showing a configuration of an in-vehicle navigation device according to an embodiment of the present invention.
[0016]
Reference numeral 1 denotes a DVD in which map data and the like are stored. Reference numeral 3 denotes an operation unit provided with various operation buttons and the like for operating the navigation apparatus body 10 to be described later. Reference numeral 4 denotes an image recognition apparatus that recognizes a specific object existing along an expressway or a general road and determines a road type. Details of the image recognition device 4 will be described later.
[0017]
Reference numeral 5 denotes a GPS receiver that receives a GPS signal sent from a GPS satellite and detects the longitude and latitude of the current position of the vehicle.
[0018]
Reference numeral 6 denotes a self-contained navigation sensor, and the self-contained navigation sensor 6 includes an angle sensor 6a such as a gyro that detects a vehicle rotation angle, and a travel distance sensor 6b that generates a pulse every certain travel distance. Reference numeral 7 denotes a display device such as a liquid crystal panel. The navigation device main body 10 displays a map around the current position of the vehicle on the display device 7, a guidance route from the departure point to the destination, a vehicle position mark, and others. The guide information of is displayed. Reference numeral 8 denotes a speaker for providing guidance information to the user by voice.
[0019]
The navigation device body 10 is composed of the following. Reference numeral 11 denotes a DVD controller that reads data from the DVD 1. Reference numeral 13 denotes an interface connected to the operation unit 3, 14 denotes an interface connected to the image recognition device 4, 15 denotes an interface connected to the GPS receiver 5, and 16 denotes an interface connected to the self-contained navigation sensor 6.
[0020]
Reference numeral 17 denotes a buffer memory that temporarily stores map data read from the DVD 1. Reference numeral 18 denotes a control unit constituted by a microcomputer. The control unit 18 detects the current position of the vehicle based on information input from the interfaces 15 and 16, reads predetermined map data from the DVD 1 to the buffer memory 17 via the DVD controller 11, Various processes such as searching for a guidance route from the departure place to the destination are performed under the search conditions set using the read map data.
[0021]
19 is a map drawing unit for generating a map image using the map data read out to the buffer memory 17, and 20 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.
[0022]
21 is a guide route storage unit that stores the guide route searched by the control unit 18, and 22 is a guide route drawing unit that draws the guide route. In the guidance route storage unit 21, all nodes of the guidance route searched by the control unit 18 are stored from the departure point to the destination. The guide route drawing unit 22 reads the guide route information (node sequence) from the guide route storage unit 21, and draws the guide route with a color and line width different from those of other roads.
[0023]
An audio output unit 23 supplies an audio signal to the speaker 8 based on a signal from the control unit 18.
[0024]
An image composition unit 24 superimposes various marks and operation screens generated by the operation screen / mark generation unit 20 on the map image drawn by the map drawing unit 19 and the guidance route drawn by the guidance route drawing unit 22. Display on the display device 7.
[0025]
In the navigation device configured as described above, the control unit 18 detects the current position of the vehicle from the GPS signal received by the GPS receiver 5 and the signal input from the self-contained navigation sensor 6. Then, the map data is referred to, map matching is performed so that the vehicle travels on the road, and the detection position is corrected.
[0026]
Further, the control unit 18 reads map data around the current position of the vehicle from the DVD 1 via the DVD controller 11 and stores it in the buffer memory 17. The map drawing unit 19 generates a map image based on the map data read to the buffer memory 17 and displays a map image around the current position of the vehicle on the display device 7. Then, the vehicle position mark is superimposed on the map image displayed on the display device 7, and the vehicle position mark is moved along with the movement of the vehicle, or the map image is scrolled.
[0027]
Further, when the user operates the operation unit 3 to set the destination, the control unit 18 uses the current position of the vehicle as the starting point, and uses the map data on the DVD 1 for the route with the lowest cost from the starting point to the destination. To explore. Thereafter, the route obtained by the search is stored in the guide route storage unit 21 as a guide route, and the guide route is superimposed on the map image and displayed. And the control part 18 outputs guidance information suitably with a driving | running | working of a vehicle, and guides a vehicle to drive | work along a guidance route to the destination.
[0028]
FIG. 2 is a block diagram showing the configuration of the image recognition device 4. Reference numeral 41 denotes a camera (digital camera) that captures the scenery in front of the vehicle. For example, as shown in FIG. 3, the camera 41 is attached to the back side of a rearview mirror in a vehicle interior. The mounting location of the camera 41 is not limited to the position shown in FIG. 3, but it is necessary to mount the camera 41 in a location where the shooting of a clear image is not hindered by rain or dust.
[0029]
An image recognition processing unit 42 recognizes an image captured by the camera 41, determines the presence or absence of an object specific to the highway or a general road, and outputs the result. A determination processing unit 44 determines the road type of the road on which the vehicle is traveling based on the output of the image recognition processing unit 42.
[0030]
Reference numeral 43 denotes an image information database. The image information database 43 stores information such as pattern information of an object necessary for image recognition, an effective appearance rate necessary for determination of a road type, and the like. FIG. 4 shows an example of object information recorded in the image information database 43. In this embodiment, signs such as side walls, roadside white lines, emergency telephones, service areas, and the like are set as objects unique to the expressway, and road signs such as temporary stops, traffic lights, and the like are objects unique to general roads. There is a pedestrian crossing. In the image information database 43, in addition to the pattern information (including color information) of these objects, the appearance frequency, the effective appearance rate, the appearance position on the image, the road attribute, and the priority are recorded.
[0031]
Reference numeral 45 denotes a camera control unit that drives the camera 41 based on movement distance information input from the navigation apparatus body 10. The moving distance information is also supplied to the image recognition processing unit 42 via the camera control unit 45.
[0032]
Hereinafter, the operation of the image recognition device 4 will be described.
[0033]
First, when the camera 41 is attached to the vehicle, the information regarding the attachment position of the camera 41 to the image recognition processing unit 42, that is, the ground height of the camera 41, the distance from the front end of the vehicle, the distance from the left and right ends of the vehicle, the depression angle of the camera 41 and Enter the angle of view. These pieces of information are used when the image recognition processing unit 42 normalizes the image. The image normalization process is a process for correcting an image captured by the camera 41 so that the same image is obtained even if the attachment position of the camera 41 is different.
[0034]
Information on the mounting position of the camera 41 is input by operating the operation unit 3 while viewing a predetermined screen displayed on the display unit 7 when the image recognition device 4 is connected to the navigation device body 10.
[0035]
The camera control unit 45 drives the camera 41 every time the vehicle moves a certain distance (for example, 5 to 10 m) based on the moving distance information input from the navigation device body 10. The image recognition processing unit 42 normalizes the image input from the camera 41 using the attachment position information of the camera 41. Thereafter, the image recognition processing unit 42 refers to the information recorded in the image information database 43 to check whether the image captured by the camera 41 has an object specific to the highway or the general road. For example, when determining the presence or absence of a side wall, the side wall is determined when a predetermined black or gray pattern is detected on the left side of the image. When determining the presence or absence of a pedestrian crossing, it is detected as a pedestrian crossing when a predetermined white pattern is detected below the image.
[0036]
The image recognition processing unit 42 determines the presence / absence of an object for all of the objects shown in FIG. 4 for each image captured by the camera 41. In FIG. 5, an example of the analysis result of the presence or absence of each target object is shown. In the example shown in FIG. 5, it is determined that there is a side wall and a distance display board. This determination result is output to the determination processing unit 44.
[0037]
Note that the image recognition processing unit 43 does not determine the presence / absence of an object based on only one image, but recognizes the object while considering the movement distance with a plurality of continuous images. Therefore, for example, when an emergency telephone is recognized at a certain point in time and an emergency telephone is recognized at a position separated by an amount corresponding to the moving distance of the vehicle in the next image, it is recognized as the same emergency telephone.
[0038]
The determination processing unit 44 accumulates the result of detection presence / absence for each object input from the image recognition processing unit 42 for a certain amount of images (for example, for 10 to 100 images). And the determination process part 44 calculates an appearance rate for every target object, as shown in FIG. For example, when a side wall is detected in 99 images out of 100 images, the appearance rate of the side wall is 99%. In the example shown in FIG. 6, the appearance rate of the side walls is 99%, the appearance rate of the distance display board is 80%, the appearance rate of the emergency call is 1%, and the appearance rate of other objects is 0%.
[0039]
Thereafter, the determination processing unit 44 compares the appearance rate of each object calculated by the calculation with the effective appearance rate for each object recorded in the image information database 43 (see FIG. 4). When the appearance rate obtained by the calculation is close to the effective appearance rate recorded in the image database 43, it is determined that there is the target object. In the case of the example shown in FIG. 6, as shown in FIG. 7, it is determined that there is a side wall, a distance display board, and an emergency telephone. The determination processing unit 44 determines that the appearance rate obtained by the calculation is significantly different from the effective appearance rate recorded in the image database 43, and does not use it for determination of the road type.
[0040]
Next, the determination processing unit 44 determines the type of road on which the vehicle is traveling from the results of these image processes. In this case, the detection result for each object is weighted according to the priority of the object recorded in the image information database 43. Specifically, when an object with high priority is detected, the detection result is 3 points, and when an object with low priority is detected, the detection result is 1 point. Then, the total points of the objects specific to the expressway are compared with the total points of the objects specific to the general road, and when the difference is larger than a specific value (for example, 5 points), the reliability of the determination result Is determined to be high. In the example shown in FIG. 7, the total number of objects unique to the highway is 7, and the total number of objects unique to the general road is 0. Therefore, in the determination processing unit 44, the road on which the vehicle is traveling is determined. Judged as a highway. Further, the reliability of the determination result is high. The determination processing unit 44 outputs the determination result of these road types and the reliability thereof to the navigation device body 10.
[0041]
The control unit 18 of the navigation device body 10 determines the currently traveling road type with reference to the road type determination result output from the image recognition device 4 and its reliability.
[0042]
However, in the navigation device of the present embodiment, the road type is basically determined based on the vehicle position detected by using GPS, self-contained navigation and map matching together. The control unit 18 refers to the determination result of the road type by the image recognition device 4 only when traveling in an area where the general road and the highway are approaching, and the reliability of the determination result by the image recognition device 4 is high. In this case, the road type is determined by giving priority to the road type determination result by the image recognition device 4. When the reliability of the determination result of the image recognition device 4 is low, the control unit 18 determines the road type in consideration of the determination result of the image recognition device 4 and the route on which the vehicle has traveled so far. .
[0043]
In the above embodiment, the detection of the object specific to the expressway and the object specific to the general road by the image recognition device 4 has been described, but the image recognition device 4 may have other functions. Good. For example, the inter-vehicle distance may be detected by the image recognition device 4 and an alarm may be output via the navigation device body 10 when the inter-vehicle distance is small.
[0044]
In the above-described embodiment, the case where the image recognition device 4 is connected to the navigation device main body 10 has been described. However, even if the navigation device main body has a function of the image recognition device (except for the camera). Good.
[0045]
【The invention's effect】
As described above, according to the in-vehicle navigation device of the present invention, an image obtained by taking a picture of the scenery outside the vehicle is subjected to image recognition processing by the image recognition device, and the presence / absence of an object unique to the expressway and the unique road The road type of the road on which the vehicle is traveling is determined according to the result, so the road type of the road on which the vehicle is traveling can be determined even in areas where the highway and the general road are close to each other. More accurate judgment can be made. As a result, it is possible to prevent the wrong guidance information from being provided to the driver due to misidentification of the road type, and further safety is ensured.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an in-vehicle navigation device according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of an image recognition apparatus.
FIG. 3 is a perspective view illustrating an example of a mounting position of a camera.
FIG. 4 is a diagram illustrating an example of information on an object recorded in an image information database.
FIG. 5 is a diagram illustrating an example of an analysis result of the presence / absence of each object by an image recognition processing unit.
FIG. 6 is a diagram illustrating an example of an appearance rate for each target object calculated by a determination processing unit.
FIG. 7 is a diagram illustrating an example of a determination result of presence / absence of each object by a determination processing unit.
[Explanation of symbols]
1 ... DVD,
3 ... operation part,
4 ... Image recognition device,
5 ... GPS receiver,
6 ... Self-contained navigation sensor,
7 ... display device,
10 ... navigation device body,
11 ... DVD controller,
17 ... Buffer memory,
18 ... control unit,
19 ... Map drawing part,
20 ... operation screen / mark generator,
21 ... guide route storage unit,
22 ... guide route drawing unit,
41 ... Camera,
42. Image recognition processing unit,
43. Image information database,
44 ... determination processing unit,
45: Camera control unit.

Claims (4)

In an in-vehicle navigation device that guides a vehicle to a destination,
Map data storage means for storing map data;
Vehicle position detection means for detecting the current position of the vehicle;
An image recognition device that captures a scene outside the vehicle, performs image recognition processing on the image , and checks the presence / absence of an object specific to the expressway and the presence / absence of an object specific to a general road for each image;
Data on the current position of the vehicle is input from the vehicle position detection means, and guidance information is provided to the vehicle driver using the map data stored in the map data storage means, and the output of the image recognition device is referred to Control means for determining the type of road on which the vehicle is running,
The image recognition apparatus has a database in which priorities are set for each object unique to the expressway and each object unique to the general road, and the object detected in the image recognition process is set according to the priority. The vehicle-mounted navigation device characterized in that the reliability of the road type determination result is evaluated.   The in-vehicle navigation device according to claim 1, wherein the image recognition device captures a scene outside the vehicle every time the vehicle travels a certain distance.   The database includes data indicating an effective appearance rate for each object unique to the expressway and each object unique to the general road, and the image recognition device determines whether the expressway The presence rate of each target object is calculated by calculating the appearance rate of the target object specific to the target object and the target object specific to the general road, and comparing the calculation result with the effective appearance rate. The vehicle-mounted navigation device according to 1.   4. The method according to claim 1, wherein the control unit determines a road type in consideration of an output of the image recognition apparatus only when traveling on an area where an expressway and a general road are close to each other. The in-vehicle navigation device according to item.

Images