JP2014173956A - Route guide device and route guide program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To carry out route guidance using a landmark existing on a travelling route from an earlier point and with more accuracy.SOLUTION: Image recognition is carried out by imaging and storing imagings in front of a vehicle at prescribed intervals from a point before a guiding intersection by only a prescribed distance L1 m before a guiding intersection, comparing respective imaging images with a landmark standard template for the guiding intersection (branching point) and deeming recognized when a matching ratio P (80%) is exceeded. When the image is recognized at the matching ratio P, the image recognition is carried out at a matching ratio Q (20%) with the imaging images sequentially going back in the past as subject images and having a position farthest away from the guiding intersection be a visualized recognition distance among the imaging positions capable of being recognized by the matching ratio Q. Then, a usual voice guidance by distance and direction is carried out when no visible distance is stored to the intersection landmark information of the guiding intersection (at a first travelling). On the other hand, when the visible distance is stored, a voice guidance using the landmark is carried out when the visible distance is passed from the guiding intersection.

Description

  The present invention relates to a route guidance device and a route guidance program, and relates to travel route guidance performed using a mark present on the travel route.

2. Description of the Related Art Navigation devices that guide the current position of a vehicle or a terminal device or a route to a destination are widely used. In this navigation apparatus, a travel route to a destination searched using map data is acquired, and guidance of the travel route is performed by voice output or image display.
When voice guidance is used for driving routes, for example, it is common to guide the distance and the direction of travel, for example, “It is approximately XXm ahead”, but it is difficult to accurately grasp the distance. Also, when there are a plurality of intersections ahead, there is a problem that it is difficult to recognize either one.

Therefore, in order to guide the intersection to be changed more easily, Patent Document 1 proposes a technique for guiding using a mark present at the intersection.
FIG. 15 shows the state of route guidance using a mark.
As shown in this figure, when the landmark can be recognized by comparing the captured image obtained by capturing the front of the vehicle with the camera and the reference image for landmark recognition, the recognized landmark (for example, shell oil) is used. Then, “Pawn, shell oil is in the right direction” is guided.

However, in the prior art, image recognition is performed from a guidance intersection because of the fact that image recognition is performed with a somewhat high matching rate (for example, 80%) in order to avoid misrecognition and the accuracy of the in-vehicle camera. In many cases, it can be recognized only about 50 meters in front.
For this reason, even if the content of the voice guidance itself is easy to understand, there are many cases where the distance is already short when the image is recognized, and the guidance may not be in time.

As a countermeasure, if the landmark can be confirmed by image recognition (coincidence rate 80%), it is assumed that the landmark will exist after that, and guidance on the use of the landmark from the near side than the actually recognized distance may be considered. It is done.
However, the position where the driver can actually see the mark is closer to the front than the position where the image can be recognized with a high coincidence rate (the distance from the guide intersection). It is unknown whether it is visible.

Japanese Patent No. 3399506

  It is an object of the present invention to more reliably perform route guidance using a mark present on a travel route from the near side.

(1) In the first aspect of the present invention, the travel route acquisition means for acquiring the travel route, the current position acquisition means for acquiring the current position of the vehicle, and the travel route on the travel route existing ahead of the acquired current position. A guide point acquisition means for acquiring the guide points of the vehicle, an imaging means for imaging the front of the vehicle, a captured image storage means for sequentially storing captured images captured by the imaging means, and a travel route corresponding to the acquired guide points Guidance means P for obtaining the mark information of the mark existing at the guide point on the acquired travel route, the mark information acquiring means for acquiring the mark information of the mark existing in the acquired travel route, and the presence rate of the mark using the captured image The mark recognition means for recognizing the mark present at the guide point, and the captured image stored in the captured image storage means are sequentially applied to the target image in a direction retroactive from the point where the mark can be image-recognized. And using the target image, the retrospective image recognition means for recognizing the image of the landmark at a coincidence rate Q lower than the coincidence rate P and corresponding to a distance visible by the driver, and the retrospective image The recognizable means outputs the distance to the point farthest from the guide point as the visible distance among the points where the mark can be recognized, and outputs the visible distance information for storage in the mark information of the mark existing at the guide point. And the guide means is provided on the condition that the mark information of the mark existing at the guide point is passed through a point corresponding to the viewable distance depending on whether or not the viewable distance exists. There is provided a route guidance device characterized by performing guidance using
(2) In the invention described in claim 2, the mark recognition means performs mark image recognition when there is no visible distance in the mark information of the mark existing at the guide point. A route guidance apparatus according to claim 1 is provided.
(3) In the invention according to claim 3, the mark standard image acquiring means for acquiring a mark standard image corresponding to the mark existing at the guide point on the travel route is provided, and the mark recognizing means is a captured image. And the standard image of the mark are recognized by the matching rate P, and the retrospective image recognition means performs the image recognition at the matching rate Q by comparing the retrospective target image and the recognized target image. The route guidance device according to claim 1 or 2, wherein recognition is performed.
(4) In the invention according to claim 4, the guide means uses, as a target image, a picked-up image picked up by the image pickup means when a visible distance exists in the mark information of the mark present at the guide point. 4. The route guidance according to claim 1, wherein guidance using the landmark is performed on the condition that an image of the landmark can be recognized with the matching rate Q. 5. Providing equipment.
(5) In the invention according to claim 5, when the visible distance exists in the mark information of the mark existing at the guide point, the guide means is based on the passage of the point corresponding to the visible distance. The guidance according to any one of claims 1 to 4, wherein guidance using the mark is performed, and guidance using the distance to the guidance point is performed when no visible distance exists. The route guidance device described in 1. is provided.
(6) The invention according to claim 6 is characterized in that the guide means guides by replacing “distance” in the guide using the distance with the “name” of the mark as the guide using the mark. A route guidance device according to any one of claims 1 to 5 is provided.
(7) In the invention according to claim 7, the landmark information acquisition means acquires the landmark information from an information center by communication, and the visible distance output means adds landmark information to the landmark information existing at the guide point. The route guidance device according to any one of claims 1 to 6, wherein the corresponding visible distance is transmitted to the information center by communication.
(8) In the invention according to claim 8, on the travel route, a travel route acquisition function for acquiring a travel route, a current position acquisition function for acquiring the current position of the vehicle, and the acquired current position ahead. A guide point acquisition function for acquiring a guide point of the vehicle, a captured image storage function for sequentially storing captured images captured by an imaging means for capturing the front of the vehicle, and a guide function for guiding a travel route corresponding to the acquired guide point The guide point information acquisition function for acquiring the mark information of the mark existing at the guide point on the acquired travel route, and the coincidence rate P for checking the presence of the mark using the captured image. A mark recognition function for recognizing a mark existing in the image and a captured image saved by the captured image saving function are sequentially acquired as target images in a direction retroactive from the point where the mark can be recognized. Using the target image, the retrospective image recognition function for recognizing the landmark with a coincidence rate Q lower than the coincidence rate P and corresponding to the distance visible by the driver, and the retrospective image recognition function, A visible distance output function for outputting the distance from the guide point that is farthest to the point farthest from the guide point as a visible distance to the mark information of the mark existing at the guide point for storage. A route guidance program to be realized, wherein the guidance function is based on whether or not there is a viewable distance in the mark information of the mark existing at the guide point. A route guidance program characterized by the function of performing guidance using the mark is provided.

  According to the present invention, when there is a viewable distance in the mark information of the mark present at the guide point on the travel route, the guide using the mark is performed on condition that the point corresponding to the viewable distance passes. The route guidance using the mark present on the travel route can be performed more reliably from the near side.

1 is a system configuration diagram of a navigation device to which the present embodiment is applied. It is explanatory drawing showing the state of a certain intersection, and the memory content of the intersection mark information. It shows a functional configuration when performing guidance processing using a mark by an information processing device in a navigation device. It is explanatory drawing showing the route guidance in case the visual recognition distance with respect to the mark of a guidance intersection does not exist. It is explanatory drawing showing the recognition process of the mark by the image recognition part, and a retroactive image process. It is explanatory drawing showing the route guidance in case the visual recognition distance with respect to the guidance intersection Z exists by retroactive image recognition. It is a flowchart showing route guidance processing. It is a flowchart showing the content of the intersection guidance process. It is explanatory drawing explaining the specific method in case a some mark exists in a guidance intersection. It is a flowchart showing the detail of the retroactive image recognition process. It is explanatory drawing showing transition of each template used in retroactive image recognition. It represents about the voice guidance in a modification. It is explanatory drawing of the modification which made the unit to go back into two steps. It is a flowchart of the retroactive image recognition process of the modification which made the retroactive unit into two steps. It is explanatory drawing showing the route guidance using the conventional mark.

DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of a route guidance device and a route guidance program according to the present invention will be described in detail with reference to FIGS.
(1) Outline of Embodiment It is determined whether or not there is a mark at the guidance intersection from the intersection mark information. When there is a mark, imaging in front of the vehicle from the guidance intersection (branch point) by a predetermined distance L1m in front. Are captured at predetermined intervals, and each image is stored.
Image recognition is performed by comparing the feature amount extracted from each captured image from the distance L1 m before with a mark standard template (mark mark standard image as a reference image for recognizing the mark) for the guidance intersection (branch point). I do. In this embodiment, in order to confirm and recognize the presence of the mark more accurately and to ensure recognition accuracy, the image recognition is performed when a predetermined high matching rate P (pattern matching rate, for example, 80%) is exceeded. It is supposed to have been made.

If the mark can be recognized with the matching rate P, the feature amount is extracted retroactively from the previously captured image, the image is recognized with the matching rate Q (<P), and the recognition is performed with the matching rate Q. Of the imaged points of the captured image, the point farthest from the guidance intersection is stored in the intersection mark information as a visible distance (return image recognition).
Here, the coincidence rate Q is a value obtained from the coincidence rate when an image is recognized using an image captured at a position where the driver can actually confirm the mark visually. The coincidence rate Q is selected in the range of 10% to 30%, and is 20% in the present embodiment.

When the travel route guidance is performed, when the visible distance is not stored in the intersection mark information corresponding to the next guidance intersection to be guided, that is, during the first travel, normal voice guidance based on the distance and direction is performed. Further, as described above, the recognizable distance is obtained by performing the image recognition of the mark with the high matching rate P and the retroactive image recognition with the low matching rate Q, and is stored in the intersection mark information.
On the other hand, when the visible distance is stored in the intersection mark information, voice guidance using the mark is performed when the visible distance has passed from the guidance intersection.

  In the present embodiment, a travel route is searched by a navigation device, intersection mark information existing on the searched route is acquired from a server of an information processing center (hereinafter simply referred to as an information processing center), and the intersection mark information can be visually recognized. When the distance does not exist, the viewable distance obtained by the retrospective image recognition is uploaded to the information processing center, but the navigation device performs all of the travel route search, intersection mark information storage, and viewable distance storage. May be. In addition, the information processing center performs all of the travel route search, intersection mark information storage, and viewable distance storage, and the navigation device obtains necessary data (travel route and intersection mark information) from the information processing center for determination. The visible distance may be increased to the information processing center.

(2) Details of Embodiment FIG. 1 is a system configuration diagram of a navigation apparatus to which this embodiment is applied.
The navigation device is mounted on a vehicle and includes a current position detection device 10, an information processing device 20, an input / output device 40, an information storage device 50, and a communication control unit 60, as shown in FIG.

The current position detection device 10 is configured as follows.
The azimuth sensor 12 is a means for detecting an angle that has changed relative to a reference angle (absolute azimuth). In this embodiment, a gyro sensor that detects a change in angle using an angular velocity is used. Yes. Note that an optical rotation sensor attached to the rotating part of the handle, a rotary resistance volume, or an angle sensor attached to the wheel part may be used. Further, the direction sensor 12 may be, for example, a geomagnetic sensor that detects in which direction the vehicle is located from detection in the N direction based on a magnet, and may be a means for detecting an absolute direction.

The distance sensor 13 is a means that can measure the moving distance of the vehicle, and for example, a sensor that detects and counts the rotation of a wheel or a sensor that detects acceleration and integrates it twice.
The GPS (Global Positioning System) receiver 14 is a device that receives a signal from an artificial satellite, and includes various signals such as signal transmission time, position information of the receiver, moving speed of the receiver, and traveling direction of the receiver. Information can be obtained.

  The information processing device 20 performs calculation and control based on information input from the current position detection device 10 and the input / output device 40 and information stored in the information storage device 50, and displays the calculation result on the display 42, the speaker 44, and the like. It is a means to control to output to the output means.

The information processing apparatus 20 is configured as follows.
The central processing unit (CPU) 21 performs overall calculation and control of the entire navigation device.
The ROM 22 is a program related to various navigations such as a search process for a travel route to a destination, an intersection (branch point) guide process using a landmark according to the present embodiment (including a landmark image recognition process and a retrospective image recognition), and various other programs. The program is stored. The ROM 22 may be divided into a first ROM and a second ROM, and a navigation program related to voice guidance may be stored in the second ROM, and other programs may be stored in the first ROM. These programs may be stored in the information storage device 50 instead of the ROM 22.
The sensor input interface 23 is a means for receiving information from the current position detection device 10.

  The RAM 24 stores information input by the user such as destination information and passage point information input by the input device 41, and results calculated by the CPU 21 based on the user input information and route search It is a storage means for storing the map result read or the map information read from the information storage device 50. The RAM 24 stores a landmark template image and a captured image as a work area in the landmark image recognition process.

The communication interface 25 is means for inputting / outputting various information via the transmission path 45. Specifically, the GPS receiver 14, the input device 41, the imaging device 43, and the information storage device 50 are connected via the transmission path 45.
The timepiece 28 is configured by using, for example, a crystal resonator, and clocks time or provides operation timing of each part of the navigation device by oscillation.
In addition, an image processor dedicated to image processing for processing vector information processed by the CPU 21 into image information, an image memory for storing image information processed by the image processor, and audio information read from the information storage device 50 are processed. However, a voice processor dedicated to voice processing to be output to the speaker 44 may be provided.

The input / output device 40 includes an input device 41 for inputting data such as a destination, a passage point, and a search condition by a user, a display 42 for displaying an image, an imaging device 43, and a speaker 44 for outputting sound.
The input device 41 includes, for example, a touch panel, a touch switch, a joystick, a key switch, and the like.
The display 42 displays a map around the current location and a travel route to the destination.

The imaging device 43 is a device that captures an image in front of the vehicle, and captures still images at predetermined intervals from a predetermined distance L1 m before the guidance intersection at an intersection for which a visible distance is not required. As the predetermined interval, every sm travel from a point before L1 m and every t seconds from a point before L1 m are possible, but in this embodiment, images are taken every 100 msec.
Each captured image captured by the imaging device 43 is sequentially stored in the captured image 52 of the information storage device 50 together with information on the imaging position or distance information from the guidance intersection.

The information storage device 50 is connected to the information processing device 20 via the transmission path 45.
The information storage device 50 stores map data 51, a captured image 52, a landmark standard template 53, intersection landmark information 54, audio data 55, and other data 55.
The information storage device 50 is generally composed of an optical storage medium such as a DVD-ROM, a CD-ROM, or a magnetic storage medium such as a hard disk, but various types such as a magneto-optical disk and various semiconductor memories. You may comprise with an information storage medium.
Information that needs to be rewritten may be constituted by a rewritable hard disk or flash memory, and other fixed information may be a ROM such as a CD-ROM or DVD-ROM.

The map data 51 stores map data, road data, intersection data, destination data, guidance point data, detailed destination data, and other data as various data necessary for map display, route search, and route guidance in navigation. Has been.
As map data, a national road map, a road map of each region, a house map, or the like is stored. The road map is composed of roads such as main arterial roads, expressways, narrow streets, and ground targets (facility etc.). A house map is a city map in which a figure representing an outer shape of a ground building, a road name, and the like are displayed. The narrow street is, for example, a relatively narrow road having a road width equal to or less than a predetermined value below a national road or prefectural road.
As for the map data, a map of a predetermined range including a current vehicle position and a point designated by the user is displayed on the display 42 at a predetermined scale. On this map, the current position of the vehicle and the designated point are displayed.

The road data is data regarding roads such as the position and type of each road, the number of lanes, and the connection relationship between the roads, and is composed of node data and link data. The road data is used for route search and map matching, and is also used when the searched travel route is displayed over the map data.
The node data is data representing geographic coordinate data of each node used for route search on a map.
For example, a road connection point such as an intersection is represented by a node, and a road between the connection points (that is, a non-branching area of the road) is represented by a link. In this way, it functions as route data representing the connection relation of the node data route.
In addition, for those where travel is restricted due to traffic restrictions, such as entry prohibition and one-way traffic, attributes indicating this are given to each link, but these attributes should be given to intersection nodes. Also good.
Node data is supplementary to characteristic points between intersections (for example, points at the start, middle and end of a curve, and points where the altitude changes), along with intersection nodes that are always set for each intersection. There is an auxiliary node that may be set to The intersection node includes information related to the intersection such as the geographical position coordinates and name of the intersection.

The intersection data stores intersection-related data such as intersection numbers, intersection geographical location coordinates and names, and link data connected to each intersection.
The destination data is data such as the locations and names of places and facilities that are likely to be destinations such as major tourist spots, buildings, and companies / businesses described in the telephone directory.
Guidance point data is guidance data of a point where guidance is required, such as the content of a guidance display board installed on a road and guidance of a branch point.
Detailed destination data is detailed data related to each facility, etc. stored in the destination data, for example, each tenant's data in a general building, facility parking lots, entrance / exit information, etc. are associated with each destination data. Stored separately. Note that the detailed destination data may be stored together in the destination data.

In the captured image 52, captured images captured by the imaging device 43 are sequentially stored together with the captured position information. The captured image 52 is deleted after image recognition processing and retrospective image recognition for the guidance intersection is completed, or is deleted by overwriting the captured image for the next guidance intersection.
Each captured image 52 is used as a target image to be recognized as to whether or not it matches the mark standard template 53 at the matching rate P.
Also, in retrospective image recognition in which recognition is performed with a matching rate Q retroactively to the previous captured image, the captured image immediately before the retrospective captured image (the side closer to the guidance intersection before going back) is a template. Used.

The mark standard template 53 is a reference image for recognizing the mark, and is stored together with the mark type (identification number) for each mark.
This mark standard template 53 is used when image recognition is performed with the matching rate P.

In the intersection mark information 54, information regarding the mark present at each intersection is stored.
FIG. 2 shows the state (a) of a certain intersection K001 and the stored contents (b) of the intersection mark information 54. FIG.
As shown in FIG. 2A, the intersection (crossroad) of intersection number K001 is connected to links R1, R2, R3, and R4, and has three mark types (identification numbers) at three of the four corners. It is assumed that marks A to C of M123, M001, and M531 exist.

On the other hand, as shown in FIG. 2B, the intersection mark information 54 stores an intersection number, a mark position, a mark type, an approach link, an exit link, and a visible distance.
As the intersection number, the same number as the intersection data of the map data 51 (intersection number K001 in the example of FIG. 2A) is stored. Since the travel route data searched for the route includes an intersection number, intersection mark information for the intersection to be guided is specified from the intersection number.

The mark position is information for specifying the position of the mark at the intersection, and can be specified by various methods such as specifying by the coordinate data of the mark. In the present embodiment, two link numbers facing the mark (identical to the link number stored in the intersection data) are specified.
In the road data, as the link connected to each intersection, the link for entering the intersection and the link for exiting from the intersection are specified separately. The mark position is specified by these two, but in FIG. 2, both are collectively shown for easy understanding.

As the mark type, the identification number of the mark standard template 53 corresponding to the mark existing at each mark position (the mark type M123 of the mark A, etc.) is stored.
For each intersection number, a mark position and a mark type are stored as a pair.

On the other hand, separately from the pair of the mark position and the mark type, the visible distance is stored for each set of the entrance link and the exit link for each intersection number.
The approach link and the exit link represent a route when passing through the intersection, and are a link entering the intersection and a link exiting the intersection.
The visually recognizable distance is a distance to a point farthest from the intersection among points that can be recognized with the matching rate Q in the retrospective image recognition according to the present embodiment. As shown in FIG. 2B, the visually recognizable distance is defined for each traveling direction with respect to the intersection specified by the approach link and the exit link.

  In the intersection mark information 54 shown in FIG. 2B, the visible distance to the intersection of the intersection number K001 in FIG. 2A is 200 m for the route R1-R2, 200 m for the route R2-R3, and R2 for each route. -R1 is determined to be 180 m, route R3-R2 is 150 m, route R4-R3 is 150 m, and stored.

In principle, the landmarks to be recognized in each route correspond to the landmarks facing the entrance link and the exit link.
For example, in FIG. 2 (a), the route entering from link R1 and exiting to link R2 and the route entering from link R2 which is the opposite route and exiting to link R1 face both links as a rule. The mark A of the mark type M123 is the recognition target.
There are no marks at the positions facing the links R3 and R4. The mark in such a case is determined in accordance with the priority order described later. In the route entering from the link R3 and leaving the link R4, the mark C of the mark type M531 is set in the route entering from the link R4 and leaving the link R3. A mark B of the mark type M001 is determined as a recognition target.

In the example of the intersection mark information 54 shown in FIG. 2B, the visible distance is not stored for the three routes R1-R4, R3-R4, and R4-R1. These three routes are all routes for which the mark C is a recognition target. For example, since the mark C was recently installed, it is considered that there is no vehicle that has traveled the three routes.
For this reason, when the vehicle travels for the first time on the above three routes, image recognition of the mark type M531 with a high coincidence rate P and retroactive image recognition with a low coincidence rate Q are performed, and a visible distance is obtained.

  As for the intersection mark information 54, a pair of a mark position and a mark type using the intersection number as a link key, and a pair of a guide route (entrance link-exit link) and a viewable distance using the intersection number as a link key are separately provided. You may make it preserve | save as this data.

Returning to FIG. 1, the case where the mark standard template 53 and the intersection mark information 54 are stored in the information storage device 50 has been described. However, when acquiring these data from the information processing center, either or both of them are acquired. You may make it preserve | save in RAM24.
In this embodiment, a case where both the mark standard template 53 and the intersection mark information 54 are stored in the information storage device 50 as shown in FIG. 1 will be described. However, since the intersection mark information 54 is acquired from the information processing center, it is stored in the RAM 24. You may make it preserve | save.

The voice data 55 stores voice data for voice guidance based on the navigation function (including intersection guidance using distances and landmarks) and various voice data such as voice data for operation assistance.
The other data 56 includes, for example, image data of photographs showing places where visual indications such as various facilities and sightseeing spots or major intersections are required, and voice when guiding a set travel route by voice. Data etc. are stored.

The communication control unit 60 transmits / receives data to / from an external device, and in this embodiment, the corresponding intersection mark information is acquired from the information processing center by transmitting the intersection number of the guidance intersection on the travel route to the information processing center. . The acquired intersection mark information is stored in the intersection mark information 54 of the information storage device 50.
Further, in the retrospective image recognition process of the present embodiment, when the visible distance is obtained by recognition with the low coincidence rate Q, it is stored in the area corresponding to the corresponding guide route (entry link-exit link) The obtained visible distance is uploaded from the communication control unit 60 to the information processing center.

FIG. 3 shows a functional configuration when the information processing device 20 in the navigation device 1 performs guidance processing using a mark.
The CPU 21 of the information processing apparatus 20 includes a navigation program stored in the ROM 22 (a search process for a travel route to a destination, an intersection guidance process using a landmark (including a landmark image recognition process, a retrospective image recognition), and the like. ) To function as the location unit 211, the image recognition unit 212, and the guide unit 213.
The location unit 211 identifies the current position of the vehicle on the map data according to the data supplied from the current position detection device 10, in this embodiment, the current position on the travel route, and sends the vehicle position information to the guide unit 213. Supply.

The image recognizing unit 212 sequentially stores the captured images captured by the imaging device 43 in the captured image 52 based on an instruction from the guide unit 213, and uses the mark standard template 53 to determine the guidance intersection with the high coincidence rate P. Recognize the landmark image.
Further, the image recognition unit 212 searches the stored captured image 52 and performs image recognition with a low coincidence rate Q, thereby obtaining a visible distance to the mark.
The recognition result by the image recognition unit 212 is supplied to the guide unit 213.

The guide unit 213 specifies a guidance intersection to be next guided on the traveling route from the traveling route stored in the RAM 24 and the own vehicle position information supplied from the location unit 211, and voice guidance about the guidance intersection is performed. Is output from the speaker 44 and a guidance image is displayed on the display 42.
The guidance unit 213 performs voice guidance for the intersection using the mark if the visible distance is stored as the voice guidance of the identified guidance intersection with respect to the guidance route (entry link-exit link) of the guidance intersection. .
On the other hand, if the visually recognizable distance is not stored, the guide unit 213 performs voice guidance using the distance as usual, and also stores the captured image for the guidance intersection and the image recognition of the mark with respect to the image recognition unit 212. And instructing backward image recognition.

In the navigation device configured as described above, route guidance is performed as follows.
The navigation device detects the current position with the current position detection device 10, reads the map information around the current position from the map data 51 of the information storage device 50, and displays it on the display 42.
When the destination is input from the input device 41, the information processing device 20 searches for (calculates) a plurality of travel route candidates from the current position to the destination, and displays them on the map displayed on the display 42. When the driver selects any travel route, the travel route is acquired by storing the selected travel route in the RAM 24 (travel route acquisition means).

The information processing apparatus 20 transmits the current vehicle position (or input departure point) and destination to the information processing center (server), and receives a travel route to the destination searched for in the information processing center. You may make it acquire a travel route by. In this case, communication of the destination and the travel route is performed by wireless communication via the communication interface 25.
Further, a travel route from the departure place to the destination is searched using an information processing device such as a personal computer at home, etc., stored in a storage medium such as a USB memory, and acquired through the storage medium reader. You may do it. In this case, the storage medium reading device is connected to the information processing device 20 via the transmission path 45.

When the vehicle travels, route guidance is performed by tracking the current position detected by the current position detection device 10.
In the route guidance, the vehicle position on the map is specified by map matching between the road data corresponding to the acquired travel route and the current position detected by the current position detection device 10, and a map around the vehicle current position is displayed on the display 42. Then, the searched travel route is displayed on the map, and a current position mark indicating the current position of the vehicle is displayed on the map.
Also, based on the relationship between the searched travel route and the current position, the necessity of guidance is determined, that is, whether or not guidance for a travel route such as a predetermined route change point or intersection guidance is necessary when straight traveling continues for a predetermined distance or more. If necessary, display on the display 42 and voice guidance are executed.

In the navigation device of the present embodiment, intersection guidance using a mark is performed for an intersection whose course is to be changed on the travel route. Hereinafter, intersection guidance using the landmark according to the present embodiment will be described.
FIG. 4 shows route guidance when there is no viewable distance with respect to the guide intersection mark.
As shown in FIG. 4, a case where a guidance intersection Z exists on the travel route R to be guided will be described as an example.
The guiding unit 213 confirms the guidance intersection Z that is the next guidance target, and confirms from the intersection mark information 54 that there is no visible distance with respect to the travel route (right turn direction) with respect to the guidance intersection Z.

Since there is no viewable distance with respect to the travel route of the guidance intersection Z, the guidance unit 213 confirms that the current position of the vehicle has reached the distance L1m (for example, 300 m) from the guidance intersection Z. On the other hand, it instructs the determination of the visible distance by sequential storage of captured images, mark recognition with a high matching rate P, and retroactive image recognition with a low matching rate Q.
In response to this instruction, the image recognizing unit 212 performs imaging in front of the vehicle by the imaging device 43 at predetermined intervals, sequentially stores the captured images, and recognizes a mark (matching rate P) with respect to the captured image from a point at a distance L1m. Start. Details of landmark recognition processing and retrospective image recognition for the guidance intersection Z will be described later.

On the other hand, since there is no visible distance, the guide unit 213 performs intersection guidance according to distance and direction as usual, as shown in FIG.
When the guide unit 213 confirms that the current position of the vehicle has reached the point X11 at a predetermined distance (for example, 300 m) before the guidance intersection Z, “Pawn, approximately ○ m ahead, right direction” Provides voice guidance according to the direction of travel. Here, the first “pawn” is a warning sound for the driver to recognize that voice guidance is provided.
When the vehicle further travels and the vehicle reaches a point X12 at a predetermined distance (for example, 100 m) further from the guidance intersection Z, the guidance unit 213 will guide “Pawn, Soon, right”.
The above voice guidance is performed by reading the voice data 55.

FIG. 5 shows a landmark recognition process and a retrospective image process performed by the image recognition unit 212. In FIG. 5, in order to make the explanation easy to understand, the display of the intersection existing on the way to the guidance intersection Z is omitted.
In parallel with voice guidance based on distance and direction, the guide unit 213 performs mark recognition processing from the captured image from a point L1m before the guidance intersection Z. In this landmark recognition process, the landmark type of the landmark M present at the guidance intersection Z is acquired from the intersection landmark information 54, and the landmark standard template 53 (reference image) corresponding to the landmark type (identification number) is read out.
Then, it recognizes whether or not the mark standard template 53 is included in the captured image in front of the vehicle captured by the imaging device 43 (captured image around the guidance intersection Z). In this image recognition, when the matching rate between the feature quantity in the captured image and the mark standard template 53 is equal to or higher than the matching rate P, it is determined that the mark has been recognized.

As shown in FIG. 5, for example, in the image recognition started from L1m before the guidance intersection Z, when the mark cannot be recognized, the same applies to the captured image taken at a point closer to the guidance intersection Z. The mark standard template 53 is recognized.
In the example shown in FIG. 5, as indicated by “impossible”, the image recognition cannot be performed with the coincidence rate P or higher in any of the captured images at the points 200 m, 150 m, and 100 m in front, and 50 m before. It is assumed that the mark has been recognized with a matching rate P or higher for the captured image.

If the mark image can be recognized, the image recognition unit 212 performs the next retrospective image recognition.
That is, the feature amount is extracted retroactively from the point where the captured images sequentially stored from the point before L1 m can be recognized with the coincidence rate P (in the example of FIG. 5, the point 50 m before the intersection). From this, it is estimated whether it can be visually recognized by the driver (visible distance).
For example, if the match rate (pattern matching rate) exceeds 80% at a 50 m point and a mark is recognized, the captured image is taken at a position that sequentially goes back from the 50 m point (each position away from the guidance intersection). As shown in FIG. 5, it is assumed that the coincidence rate has changed to 50% at a point 100 m in front, 20% at a point 150 m, and 0% at a point 200 m.
In this case, if the coincidence rate P at the point where the driver can actually visually recognize the mark is 20%, the visible distance is 150 m before the guidance intersection.

Actually, the captured images are sequentially traced from the 50 m point, the imaging point of the captured image having a coincidence rate of P = 20% or more is examined, the distance from the guidance intersection to the farthest point is determined as the visible distance, and the intersection mark Stored in information 54.
In addition, when retrospective image recognition is performed with the matching rate P, the reference image to be compared with the feature amount extracted from the captured image is not a landmark standard template, but an image of the recognition result immediately before the backward image recognition (guide intersection side) ( Captured image). As described above, in retrospective image recognition, image recognition with a matching rate P is performed based on a mark image included in a captured image that actually exists at a guidance intersection instead of a mark standard template that is a general mark image. Thus, the recognition rate close to that of an actual driver can be obtained.

FIG. 6 shows route guidance when there is a visible distance with respect to the guidance intersection Z by retrospective image recognition.
When the visible distance to the mark existing at the guidance intersection Z is stored by the above-described retrospective image recognition, the intersection guidance using the mark is performed for the subsequent travel.
The guidance unit 213 confirms the guidance intersection Z that is the next guidance target, and confirms by the intersection mark information 54 that there is a visible distance with respect to the travel route (right turn direction) with respect to the guidance intersection Z. Check the current position, and when you reach the point X22 that is just a visible distance from the guidance intersection Z, include “Shell oil” as a landmark in the guidance voice, such as “Pawn, shell oil is in the right direction”. Provide voice guidance. In the voice guidance using the mark, when the voice data 55 is read, the explanation part of “distance” in the normal guidance using the distance is changed to the explanation of “mark”.

Next, details of the route guidance processing in this embodiment will be described.
FIG. 7 is a flowchart showing route guidance processing.
The guide unit 213 acquires a travel route to the destination, stores it in the RAM 24 (step 1), and monitors whether the travel has started (step 2).
The travel route to the destination is acquired by performing a route search to the destination input in the own device, but may be acquired from the information processing center. In this case, the destination input by the user is transmitted to the information center, and the travel route to the destination searched for in the information center is received.
Whether or not the vehicle has started traveling is determined based on whether or not the current position of the vehicle has moved. Note that it may be determined that the vehicle has started traveling even when the vehicle moves at a predetermined vehicle speed or higher.

If it is determined that the vehicle has started traveling (Step 2; Y), the guide unit 213 acquires the next guidance point existing in front of the vehicle from the traveling route stored in the RAM 24 (Step 3), and determines whether the next guidance point is an intersection. It is determined whether or not (step 4).
When the next guidance point is not an intersection (step 4; N), the guidance unit 213 performs guidance processing according to the guidance point (step 5). For example, the guidance unit 213 performs voice guidance such as “Pawn, it is a 4 km road as it is” and screen display.

On the other hand, when the next guidance point is an intersection (guidance intersection) (step 4; Y), the guide unit 213 performs an intersection guidance process described later (step 6).
Then, the guide unit 213 determines whether or not the current position of the vehicle supplied from the location unit 211 has arrived around the destination (step 7). If it has not arrived (step 7; N), step 3 Returning to, the route guidance is continued, and if it has arrived around the destination (step 7; Y), the process returns to the main routine.

FIG. 8 is a flowchart showing the contents of the intersection guidance process in step 6.
The guide unit 213 acquires the distance from the vehicle position acquired from the location unit 211 to the guidance intersection (step 11), and determines whether or not the vehicle is located within L1m before the guidance intersection and after passing the guidance intersection within L2m. Judgment is made (step 12). The distance L2m is a distance determined in consideration of the time until the completion of the retrospective image recognition, and 100 m is defined in the present embodiment.
If it is not located within the range of the distances L1m to L2m (step 12; N), the process for the guidance intersection has been completed, and the process returns to the main routine.

When it is located within the distance L1m to L2m (step 12; Y), the guide unit 213 determines whether or not the intersection mark information of the guidance intersection has been acquired (step 13).
If the intersection mark information has not been acquired (step 13; N), the guide unit 213 transmits the intersection number of the guide intersection to the information center via the communication control unit 60, thereby obtaining the intersection mark information from the information center. It is acquired and stored in the intersection mark information 54 of the information storage unit 50 (step 14).

Next, the guide unit 213 determines whether or not the voice guidance for the guidance intersection has been completed (step 15). If the voice guidance has been completed (step 15; Y), the process proceeds to step 21.
On the other hand, if the voice guidance has not been completed (step 15; N), the guidance unit 213 confirms whether there is a landmark at the guidance intersection with the intersection landmark information 54 of the information storage unit 50 (step 16). (Step 16; N) Return to the main routine.

If there is a mark at the guidance intersection (step 16; Y), the guidance unit 213 determines whether or not a visible distance is stored in the intersection mark information 54 for the guidance route (entrance link-exit link). (Step 17). If the viewable distance is not stored (step 17; N), the guide unit 213 performs voice guidance based on the distance and direction as described with reference to FIG. 4 (step 18).
On the other hand, when the visible distance is stored (step 17; Y), the guide unit 213 determines whether the distance from the vehicle position to the guidance intersection is equal to or less than the visible distance (step 19). If it is not less than the visible distance (step 19; N), the process proceeds to step 21.
If the distance is less than the visible distance (step 19; Y), the intersection guidance using the mark is performed as described in FIG. 6 (step 20). In this case, for example, assuming that the vehicle travels 1 m during the call timing of processing, usually, voice guidance using a mark is performed within a range of about 1 m from the visible distance.

Next, the guide unit 213 determines whether or not the mark has been image-recognized (completed) for the guidance intersection (step 21).
When the landmark has already been recognized (step 21; Y), there is a viewable distance for the guide intersection mark, and there is no viewable distance, but the viewable distance has been calculated by retrospective image recognition. More specifically, Step 19; the case where the transition is made from N, the case where the transition is made from Step 20, and the case where Step 20 is executed in the previous loop and the transition is made from Step 15 are applicable. .
On the other hand, when the landmark has not been image-recognized (step 21; N), there is no viewable distance with respect to the landmark at the guidance intersection. This corresponds to the case where step 18 is executed in the previous loop and the process proceeds from step 15 this time.

If the landmark has been image-recognized (step 21; Y), the guide unit 213 returns to the main routine.
On the other hand, if the mark has not been image-recognized (step 21; N), the guide unit 213 sets the image recognition threshold to the coincidence rate P (step 22), and instructs the image recognition unit 213 to perform image recognition. Then, the image recognition unit 212 recognizes the image of the mark (step 23).

In the image recognition of the landmark, the image recognition unit 212 identifies the landmark type (identification number) of the landmark to be recognized from the guide route (entrance link-exit link) based on the intersection landmark information 54, and the landmark corresponding to the landmark type The standard template 53 is read and stored in the RAM 24.
When there are a plurality of landmarks for the guidance intersection, the priorities for determining the positions of the landmarks to be recognized are determined in advance in the following order (1) to (4) according to the travel route. .
(1) Position on the near side in the direction to turn left or right at the guidance intersection (2) Position on the far side in the direction to turn left or right at the guidance intersection (3) Position on the near side in the direction opposite to the direction to turn left or right at the guidance intersection ( 4) Position on the far side in the direction opposite to the direction of turning left or right at the guidance intersection

FIG. 9 is a diagram for explaining a specifying method in the case where a plurality of landmarks exist at the guidance intersection.
According to the above priority order, when a travel route R that turns right at the guidance intersection is set as shown in FIG. 9, the mark present at the position A on the right front side is determined according to the priority order (1).
If there is no mark at position A, the mark present at position B on the far right side is determined according to priority (2).
If there is no mark at position A or B, the mark present at position C on the left front is determined according to priority (3).
If there is no mark at any of the positions A to C, the mark present at the position D on the far left side is determined according to the priority (4).
Further, when a travel route opposite to the travel route R shown in FIG. 9 is set, the positions are determined in the order of position A, position C, position B, and position D.

The image recognizing unit 212 can recognize the mark depending on whether or not the feature amount extracted from the captured image captured by the imaging device 43 and the mark standard template 53 corresponding to the determined mark position are equal to or higher than the matching rate P. It is determined whether or not (step 24).
When the coincidence rate is less than P and the mark cannot be recognized from the captured image (step 24; N), the process returns to the main routine. In this case, when the intersection guidance processing is performed in the next loop, the matching rate with respect to the next captured image (captured image captured at a position closer to the guide image) after passing through Step 21; N and Step 22 Image recognition at P is performed.

When the image of the mark is recognized (step 24; Y), the image recognition unit 212 performs a retrospective image recognition process (step 25, details will be described later).
Then, the guide unit 213 stores the viewable distance obtained by the image recognition unit 212 through retrospective image recognition in the guide route (entrance link-exit link) corresponding to the intersection number in the intersection mark information 54, and the intersection number and the guide route. The viewable distance obtained together with (entrance link-exit link) is increased to the information center (server) (step 26), and the process returns to the main routine.

FIG. 10 is a flowchart showing details of the retrospective image recognition process.
The image recognizing unit 212 changes the current target image, that is, the captured image that is the recognition target in the current image recognition, from the mark standard template that has used the image recognition template at the matching rate P to the template. (Step 251).
FIG. 11 shows the transition of each template used in retrospective image recognition.
As shown in FIG. 11A, for example, when the landmark can be recognized at the point 50 m before the guidance intersection with the matching standard P with the landmark standard template, in step 251, it is used for this image recognition. An image obtained by extracting the mark portion from the captured image is changed to a template.

Next, in steps 252-1 to 252-2, the image recognition unit 212 performs image recognition on each retrospective target image (a captured image that is a recognition target), and the image recognition distance (distance from the guidance intersection) is L3m. Steps 253 to 258 are performed in a loop until the value reaches.
First, the image recognition unit 212 returns a target image for image recognition, that is, a captured image to be recognized, by L4m from the previous target image (step 253). That is, the image recognizing unit 212 reads a captured image captured at a point closest to a point returned L4m from the point where the previous target image was captured from the captured image 52 of the information storage device 50 as the current target image. When the point returned by L4m is farther from the guidance intersection than L3m, the image captured at the point closest to L3m is set as the current target image.
The distance L4 is less than L3, and in this embodiment, L3 is set to 300 m and L4 is set to 50 m as an example.

Then, the image recognizing unit 212 sets the retroactive image recognition threshold from P% to Q% (match rate Q) (step 254). In the present embodiment, the coincidence rate Q is set to 20%.
Next, the image recognition unit 212 performs an image recognition process for the intersection mark (step 255), and whether or not the recognition has been completed, that is, whether or not the changed template and the current target image are equal to or higher than the matching rate Q. Is determined (step 256).
In the example shown in FIG. 11, as shown in FIG. 11B, the captured image before 50 m is changed to the current template (step 251), and the captured image obtained by returning L4m (= 50 m + 50 m = 100 m) is the current target image. (Step 253). Then, it is determined whether or not the coincidence ratio between the current target image (captured image before 100 m) and the template (captured image before 50 m) is Q = 20% or more. In the example of FIG. 11B, since the coincidence rate is 20% or more at 50%, it can be recognized.

When the image can be recognized (step 256; Y), the image recognition unit 212 sets the visible distance = current distance (step 257). In the example of FIG. 11B, since it is recognized, the visible distance = current distance = 100 m.
The image recognition unit 212 changes the image recognition template to the current target image (step 258). That is, the image recognition unit 212 changes the captured image (target image) that is the recognition target in the current image recognition to the template. In the example of FIG. 11B, the template is changed to a captured image 100 m before, which is a recognition target image.

In step 256, when the image recognition with the coincidence rate Q cannot be performed (step 256; N), the image recognition unit 212 exits from the processing loop (steps 251-1 to 251-2) and returns to the main routine. To do.
In this case, the current distance stored when the previous loop is recognized is the visible distance in the retrospective image recognition.
In the example of FIG. 11, as shown in (d), it is determined that the matching rate with the template (captured image before 150 m) with respect to the captured image before 200 m is 10% and cannot be recognized. In this case, the loop immediately before image recognition, the current distance 150 m in FIG. 11B is the visible distance.

When the processing from step 253 to step 258 is looped until the image recognition distance L3m is reached, the process returns to the main routine.
In this case, the current distance (the distance to the imaging position of the target image) in the last loop is the visible distance.

Although one embodiment of the navigation device 1 of the present invention has been described above, the present invention is not limited to the described embodiment, and various modifications can be made within the scope described in each claim. .
For example, in the embodiment described above, when the visible distance with respect to the guide intersection mark is present in the intersection mark information 54, as shown in FIG. 19; Y), the case where voice guidance using a mark (step 20) is performed has been described.
On the other hand, in this modification, when a visible distance exists in the intersection mark information 54 of the information storage unit 50, the image recognition threshold is lowered to the coincidence rate Q, the image recognition is performed, and the existence can be confirmed. In addition, voice guidance is performed by means of landmarks.
FIG. 12 shows voice guidance in this modification.
In the example shown in FIG. 12, the image recognition of the mark is started with the coincidence rate Q (20%) with the mark standard template, with the captured image taken at the position X42 of the visible distance by the vehicle as the target image. In the example of FIG. 12, since the coincidence rate Q could not be recognized at the position X42, the captured images sequentially picked up are recognized, and since the mark can be recognized at the position X43, voice guidance using the mark is output. Shows the state.

In this modification, a landmark standard template is used as a template to be compared with the target image. However, a template used in the retrospective image search, that is, a target image actually captured may be used as the template.
In addition, when both of the mark standard template and the target image are used as templates and the mark image is recognized with the matching rate Q with respect to one of them, voice guidance using the mark may be performed.
When a target image is used as a template, a captured image (target image 50 m before in FIG. 11A) used when the mark is recognized with the matching rate P using the mark standard template is used. Alternatively, the captured image (target image before 150 m in FIG. 11C) used when determining the visible distance may be used.
In any case, when using the actually captured target image as a template, the guide unit 213 stores the template target image including the target image as the template in step 26 and uploads it to the information center. To do.

In the embodiment described above, in the voice guidance based on the distance and direction with respect to the guidance intersection, as shown in FIG. 4, “Pawn, soon, right direction” at a predetermined distance (for example, 100 m) from the guidance intersection. So that the last voice guidance.
Therefore, as another modification, the last voice guidance may be the target of intersection guidance using a mark.
For example, as shown in FIG. 12, when the vehicle is in front of a predetermined intersection just before the guidance intersection (for example, 100 m before), the guidance is given as “Pawn, shortly, the shell oil is in the right direction”. May be.

In the retrospective search described with reference to FIG. 10, a case has been described in which a captured image corresponding to a position retroactive by a predetermined distance L4m is used as a target image for which a landmark is to be recognized. In this case, the value of L4 is increased. In this case, the number of times of recognition is reduced and the processing is quickened. However, since the unit to be traced is rough, there is a possibility that the driver may be in front of the distance where the landmark can actually be visually recognized.
On the other hand, if the value of L4 is decreased, a visible distance closer to the actual visible distance by the driver can be obtained, but the number of recognition processes (the number of loops in steps 252-1 to 253-2) increases. There is a problem that recognition time becomes long.
Therefore, in this modified example, the retroactive unit is divided into two stages of large and small.
That is, as the predetermined distance L4, two types of a large value L5 (for example, 50 m) and a minimum unit Min (for example, 1 m) are set, and at the beginning, it goes back in units of L4 = L5. This is one that goes back in increments of L4 = Min.

FIG. 13 is an explanatory diagram when the retroactive unit is in two stages.
Explaining this modified example with reference to FIG. 13, if the mark standard template and the matching rate P (for example, 80%) can be recognized at a point Xm before the guidance intersection Z (step 24; Y), the image recognition unit 212 , L4 = L5m (for example, 50 m), the image captured in correspondence with the position that is traced back by L5m (for example, 50 m) is used as the target image, and image recognition is performed with a matching rate Q (for example, 20%) with the template. Image recognition is sequentially performed on the target images.

If image recognition with the coincidence rate Q cannot be performed (when the coincidence rate is less than 20%), the process returns to the immediately preceding target image position where the image could be recognized, and the distance L4 retroactively from L5 to Min (for example, 1m), and the image recognition corresponding to the template and the coincidence rate Q is performed with the captured image corresponding to the position traced back in detail as the target image. Thereafter, if the image of the mark can be recognized, image recognition is sequentially performed on the target image that is further back by Min.
Then, when image recognition with the matching rate Q cannot be performed on the target image that is traced back by Min as the minimum unit (when the matching rate is less than 20%), The distance from the guidance intersection to the imaging position is the visual recognition distance.
The template used in this modification example uses the template described in the embodiment or its modification.

FIG. 14 is a flowchart of a retroactive image recognition process according to a modified example in which the retroactive unit is divided into two stages.
In the embodiment, the same steps as those described in FIG. 10 are denoted by the same step numbers, and description thereof will be omitted as appropriate.
The image recognizing unit 212 changes the current target image, that is, the captured image that is the recognition target in the current image recognition, from the mark standard template that has used the image recognition template at the matching rate P to the template. At the same time, the retroactive distance L4 is set to L4 = L5 (step 251b).
Thereafter, the image recognition unit 212 repeats the loop from step 252-1 to step 252-2 as described in the embodiment. In this loop, as in the embodiment, the landmark image recognition with the matching rate Q is performed with the captured image corresponding to the position retroactive by a large retrospective distance L4 = L5 as the target image.

Then, if the landmark image recognition with the coincidence rate Q cannot be performed in this loop (step 256; N), the image recognition unit 212 shifts to finely retrospective image recognition according to the present embodiment.
In other words, the image recognition unit 212 sets L4, which is a retroactive unit, as the minimum unit Min (step 261).
Then, a captured image corresponding to a point returned from the current viewable distance by L4m = Min (away from the guidance intersection) is acquired as a target image (step 262). The current viewable distance corresponds to the current distance saved in step 257 in the loop immediately before step 256; N.

  Next, the image recognition unit 212 compares the target image with the template and performs image recognition processing of the intersection mark (step 263). The template used here is the template used before exiting the loop of step 252-1 to step 252-2 in the first process, and the template after that is the template changed in step 266 described later.

Then, the image recognizing unit 212 determines whether or not the mark image has been recognized, that is, whether or not the template and the current target image have a matching rate Q or more (step 264).
When the image can be recognized (step 264; Y), the image recognition unit 212 sets the visible distance = current distance (step 265).
The image recognition unit 212 changes the image recognition template to the current target image (step 266). That is, the image recognition unit 212 changes the template to a captured image (target image) that is a recognition target in the current image recognition.

The image recognition unit 212 further returns the target image for image recognition by L4 = Min (step 267), returns to step 263, and similarly repeats steps 263 to 267.
If the image cannot be recognized in step 264 (step 264; N), the process returns to the main routine. The visible distance in this case is the visible distance stored in step 265 as a result of the previous image recognition (step 263) = current distance.

In the embodiment and the modification described above, the navigation apparatus 1 mounted on the vehicle has been described as an example. However, a mobile phone equipped with the navigation program (function) described above or a personal digital assistant (PDA: Personal Digital Assistant, PND) : Personal Navigation Device), a personal computer may be used as a navigation device in a vehicle.
In addition, when a user carries and uses each device such as a mobile phone, a moving body that changes to a vehicle may be applied as the user.

  Further, in the present embodiment and the modification, the description has been made on the guide intersection on the travel route. However, the present invention is not limited to the intersection, and guides a specific branch point existing on the travel route or guides straight ahead. It is also possible to apply to this.

DESCRIPTION OF SYMBOLS 10 Current position detection apparatus 20 Information processing apparatus 40 Input / output apparatus 43 Imaging apparatus 44 Speaker 50 Information storage apparatus 51 Map data 52 Captured image 53 Mark standard template 54 Intersection mark information 55 Voice data 60 Communication control section 211 Location section 212 Image recognition section 213 Guide

Claims (8)

Traveling route acquisition means for acquiring a traveling route;
Current position acquisition means for acquiring the current position of the vehicle;
Guidance point acquisition means for acquiring a guide point on the travel route existing ahead of the acquired current position;
Imaging means for imaging the front of the vehicle;
Captured image storage means for sequentially storing captured images captured by the imaging means;
Guidance means for guiding a travel route corresponding to the acquired guidance point;
Mark information acquisition means for acquiring mark information of a mark present at a guide point on the acquired travel route;
Mark recognition means for recognizing an image of a mark present at the guide point at a matching rate P for confirming the presence of the mark using the captured image;
The captured image stored in the captured image storage means is sequentially acquired as a target image in a direction going back from the point where the mark can be recognized, and the target image is used. The target image is lower than the matching rate P and visually recognized by the driver. Retrospective image recognition means for recognizing the image of the landmark with a matching rate Q corresponding to a possible distance;
The backward image recognition means outputs the distance to the point farthest from the guide point as a visible distance among points where the mark can be recognized, and outputs the mark information of the mark existing at the guide point for storage. A possible distance output means,
The guide means performs guidance using the mark on the condition that the mark information of the mark existing at the guide point passes through the point corresponding to the visible distance depending on whether or not the visible distance exists. ,
A route guidance apparatus characterized by that. The landmark recognition means performs image recognition of the landmark when there is no visible distance in the landmark information of the landmark existing at the guide point.
The route guidance apparatus according to claim 1. A mark standard image obtaining means for obtaining a mark standard image corresponding to a mark present at the guide point on the travel route;
The landmark recognition means performs image recognition at the matching rate P by comparing the captured image with the landmark standard image,
The retroactive image recognition means performs image recognition with the matching rate Q by comparing the retrospective target image with a recognized target image.
The route guidance device according to claim 1 or 2, characterized by the above. The guide means can recognize an image of the mark at the matching rate Q using the picked-up image picked up by the image pickup means as a target image when the mark information of the mark existing at the guide point has a visible distance. As an additional condition, guide using the landmarks,
The route guidance device according to claim 1, claim 2, or claim 3.   The guide means, when there is a viewable distance in the mark information of the mark present at the guide point, performs guidance using the mark on condition that the point corresponding to the viewable distance passes, and the viewable distance 5. The route guidance device according to claim 1, wherein when there is not, guidance using a distance to the guidance point is performed. The guidance means guides the “distance” in the guidance using the distance by replacing the “name” of the landmark as guidance using the landmark,
The route guidance device according to any one of claims 1 to 5, wherein the route guidance device is characterized by that. The landmark information acquisition means acquires the landmark information from an information center by communication,
The visually recognizable distance output means transmits to the information center by communication as the visually recognizable distance corresponding to the landmark information of the landmark existing at the guide point.
The route guidance device according to any one of claims 1 to 6, wherein the route guidance device is characterized by that. A travel route acquisition function for acquiring a travel route;
A current position acquisition function for acquiring the current position of the vehicle;
A guide point acquisition function for acquiring a guide point on the travel route existing ahead of the acquired current position;
A captured image storage function for sequentially storing captured images captured by an imaging unit that captures the front of the vehicle;
A guidance function for guiding a travel route corresponding to the acquired guidance point;
A landmark information acquisition function for acquiring landmark information of a landmark present at a guide point in the acquired travel route;
A mark recognition function for recognizing an image of a mark present at the guide point with a matching rate P for confirming the presence of the mark using the captured image;
The captured image stored by the captured image storage function is sequentially acquired as a target image in a direction going back from the point where the mark can be recognized, and the target image is used, which is lower than the matching rate P and visually recognized by the driver. A retrospective image recognition function for recognizing an image of the landmark with a matching rate Q corresponding to a possible distance;
A visual distance that is output to the mark information of the mark existing at the guide point as a viewable distance from the point where the mark can be recognized by the retrospective image recognition function as a visible distance. Possible distance output function,
A route guidance program for realizing the above on a computer,
The guidance function performs guidance using the mark on the condition that the mark information of the mark existing at the guidance point passes through the point corresponding to the visible distance depending on whether or not the visible distance exists. A route guidance program characterized by being a function.

Images