JP2009186372A - Navigation device and navigation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a navigation device and navigation method allowing a driver to accurately obtain the intersection at which he/she should turn. <P>SOLUTION: A point to be guided, for example, objects existing around a vehicle at an intersection, is photographed by an on-vehicle camera 11 and held by an image holding section 13, an image processing unit 14 extracts an object serving as a landmark (mark) from the photographed objects (stopping and parking vehicle or facilities on the road side), and an information creating-outputting section 30 creates and outputs presentation information corresponding to the feature of the object extracted as the mark. When the driver approaches the intersection or the like set as the point to be guided during route guide by a route guide arithmetic section 20, the information creating-outputting section 30 converts the presentation information into voice information and outputs from a speaker 32, and/or converts it into character-diagram information and outputs it to a display section 33. As the feature of the object extracted as the mark, at least one of the color, shape, size and advertising information of the object is extracted. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a navigation device and a navigation method.

  One of the important roles when guiding a route as navigation of a vehicle is to guide the traveling direction of the vehicle at an intersection.

In the conventional technique described in Japanese Patent Application Laid-Open No. 2004-205405 “Navigation device” in Patent Document 1, when performing guidance on where to turn, the driver knows even if the name of the intersection to be turned is transmitted. Since it is difficult, it has been proposed that facility information existing around an intersection to be turned is acquired from map information prepared in advance and used as guide information at the intersection.
JP 2004-205405 A

  The prior art described in Patent Document 1 acquires facility information existing in the vicinity of an intersection to be turned from a map database provided in advance and uses the information as guide information at the intersection. Therefore, it is not possible to deal with new facilities or facilities that have disappeared since the time when the map database was created, and old guide information may be presented. Therefore, there is a possibility that information that serves as a mark cannot be accurately grasped for a driver who is about to enter an intersection to bend.

  The present invention has been made in view of such circumstances, and provides a navigation device and a navigation method that allow a driver to accurately grasp an intersection to be bent when a vehicle is about to enter the intersection to be bent. That is the purpose.

  In order to solve the above-described problem, the present invention detects an object existing around a vehicle at a point to be guided, extracts an object as a mark from the detected objects, and corresponds to the object extracted as the mark. It is characterized by generating and outputting information presented to the driver.

  According to the navigation device and the navigation method of the present invention, since it is guaranteed that the mark object actually exists, there is an effect that the driver can accurately grasp the mark object. It is done.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of a navigation device and a navigation method according to the invention will be described in detail with reference to the drawings.

(Features of the present invention)
Prior to the description of the embodiments of the present invention, an outline of the features of the present invention will be described first. In general, information that is most easily understood as guide information at a point to be guided is not always fixed, and changes with time. The present invention actually captures a point to be guided using an in-vehicle camera, detects an object that is an appropriate landmark (marker or geographical marker) in the traveling direction of the host vehicle, Instead of information and character / graphic information, it is characterized by being presented auditorily as audio information to the driver and / or visually presented as character information and image information.

  In other words, when approaching a point to be guided, such as an intersection where a turn is to be made, information on guiding the route should be provided to the driver at the guide timing to be guided, for example, characteristics of other vehicles existing around the intersection or the roadside As information indicating the characteristics of the facility existing in the vehicle, information that is easy for the driver to visually recognize, for example, the color / shape / size of the vehicle body or the facility / signboard information written on the vehicle body / facility (character information or signboard written on the signboard) The symbol mark drawn in (1) is read and presented to the driver using language information and / or character information and graphic information.

  More specifically, the present invention calculates a recommended route (recommended route) to the destination of the own vehicle, and in a navigation device that performs route guidance along the recommended route, an object (another vehicle) existing around the own vehicle. The vehicle is equipped with an in-vehicle camera that measures and detects the distance to the detected object as an object detection means, such as a right- or left-turn at a point to be guided, such as an intersection. When performing the guide, among the objects existing near the intersection photographed by the in-vehicle camera, an object that is easy to see from the driver is extracted as a guide mark (landmark) so that the driver can easily understand. It is possible to perform a guide.

(First embodiment)
<Device configuration>
First, a first embodiment of a navigation device and a navigation method according to the present invention will be described. FIG. 1 is a block configuration diagram showing a block configuration of a first embodiment of a navigation apparatus according to the present invention. The navigation device shown in FIG. 1 is mounted on a vehicle as a device that provides route guide information to a driver, and includes at least an object recognition unit 10, a route guide calculation unit 20, and an information generation / output unit 30. ing.

  The object recognition unit 10 has a function of recognizing an object around a vehicle, that is, an object that is a landmark (landmark), and includes one or more cameras 11 mounted on the vehicle, sensors 12 mounted on the vehicle, and an image holding unit. 13, an image processing unit 14, and an object position relationship calculation unit 15. The route guide calculation unit 20 has a function of calculating and obtaining route guide information, and includes a route information comparison calculation unit 21, a route calculation unit. 22, the map database 23 is provided at least, and the information generation / output unit 30 has a function of providing the driver with route guide information obtained by calculation, and includes at least an information generation unit 31, a speaker 32, and a display unit 33. I have.

  In the configuration example shown in FIG. 1, for example, one or a plurality of on-vehicle cameras 11 are provided, and a stereo camera constituted by one or a plurality of cameras 11 is used, and is further used in combination with a vehicle-mounted sensor 12 that measures a distance. By detecting the position / features of landmarks (markers) around the host vehicle that can be used as guide marks among objects (such as other vehicles and roadside facilities) existing around the host vehicle, language information and / or Alternatively, guide information is provided to the driver by outputting from the speaker 32 or the display unit 33 instead of the character information or graphic information.

  That is, as shown in FIG. 1, for example, an object (such as another vehicle or a roadside facility) existing around the vehicle at a point to be guided, such as an intersection to be turned, is detected by an object detection unit including a camera 11, a sensor 12, and the like. Then, from the detected objects, the image processing unit 14 that provides a landmark extracting means extracts an object, that is, an object that becomes a landmark (a landmark), further extracts its features, and the route guide computing unit 20 In addition to the calculated route guide information to the destination, the features that characterize the extracted object, that is, the object that is the landmark (mark), are voiced by the information generation / output unit 30 that provides the information generation / output unit. It is converted into information and / or character / graphic information and presented to the driver as presentation information. Here, the information generation / output unit 30 is configured to output the presentation information when the vehicle approaches a predetermined point such as an intersection to be turned during the route guide.

  Next, the function of each component of the navigation device shown in FIG. 1 will be described. As described above, the camera 11 of the object recognition unit 10 is mounted on the vehicle, and is used to acquire external information around the host vehicle (information on other vehicles and roadside facilities). The sensor 12 is mounted on the vehicle. Sensor information for measuring the distance from the vehicle to the landmark (mark) of the recognized landmark (mark) from the outside information around the vehicle acquired by the camera 11 Is something to get.

  The image holding unit 13 holds image information around the host vehicle acquired by the camera 11 on a memory, and the image processing unit 14 provides a mark extraction unit. An object to be a target, that is, a landmark (landmark) is extracted from the held image information, and a feature of the object to be the target, that is, a landmark (landmark) is extracted. Here, the characteristics of the target object, that is, the landmark (landmark), are the color, shape and size of the target object (such as other vehicles and roadside facilities), characters and symbol marks written on the target object, etc. It is information that is easily noticeable including at least the signboard information.

  The object position relationship calculating unit 15 calculates the distance between the landmark (mark) extracted by the image processing unit 14 and the host vehicle using the sensor information acquired by the sensor 12.

  The route information comparison calculation unit 21 of the route guide calculation unit 20 refers to the route guidance information for navigation and detects a guide timing indicating that a guide point to be notified to the driver is approaching.

  The information generation unit 31 of the information generation / output unit 30 uses the guide timing indicating that the route information comparison calculation unit 21 is approaching the guide point to be guided, that is, the guide point indicating that the guide point is approaching. When it is detected that the route guide has been reached, the route guide calculation unit 20 uses the calculation result of the object recognition unit 10 for the route guide information (language information and graphic / character information) currently generated. Route guide information (language information and graphic / character information) relating to other landmark information newly extracted by the image processing unit 14 serving as extraction means is generated, added as presentation information, and output.

  Furthermore, information redundancy check is performed on the route guide information (language information and graphic / character information) related to the generated other landmark information, and the redundant information is deleted from the output presentation information. . In addition, in order to convert the presentation information into voice information and output it from the vehicle-mounted speaker 32, the voice information is generated using voice synthesis, and the presentation information is converted into character information or graphic information. Display information is generated for display.

  The speaker 32 of the information generation / output unit 30 is mounted on the vehicle, and is used to output language information (voice information) related to the route guide information generated by the information generation unit 31. The display unit 33 The display information (character information and graphic information) related to the route guide information generated by the information generation unit 31 is displayed on the monitor screen.

  The route calculation unit 22 of the route guide calculation unit 20 has a general navigation function, and calculates the arrival route from the current location to the destination when the destination is set. A database 23 stores map information used in the navigation function.

<Description of operation example>
Next, an example of the operation of the navigation device of FIG. 1 will be described using the flowchart shown in FIG. FIG. 2 is a flowchart showing an example of the operation of the navigation apparatus shown in FIG. 1, and shows an example of the navigation method according to the present invention.

  First, the flow of processing for extracting the characteristics of an object that becomes an object, that is, a landmark (landmark) will be described by taking as an example the case where the object, that is, the object that becomes the landmark (landmark), is a parked vehicle. Note that FIG. 2 illustrates the case where the object, that is, the object that is the landmark (landmark) is a parked vehicle, but the object that is the landmark (mark) is not limited to the parked vehicle, Any object may be used as long as it is a mark indicating the characteristics of a point to be guided such as a facility.

  Further, the flowchart of FIG. 2 shows that when the vehicle approaches a predetermined point as a point to be guided during the route guide, that is, when the guide timing at which the guide information is to be output is reached, the driver visually recognizes the route. It is activated in order to present guide information regarding an object that is an easy-to-use landmark (mark) to the driver as presentation information.

  In the flowchart of FIG. 2, first, using the vehicle-mounted camera 11, it is monitored whether or not a vehicle that can be a candidate for an object, that is, a landmark (landmark) object, exists in the traveling direction of the host vehicle. Continue (step S1). When the image processing unit 14 refers to the image information captured by the in-vehicle camera 11 and stored in the memory, when a vehicle that should be a candidate for an object, that is, a landmark (landmark), is not detected (step S1) In the case of “target candidate not detected”, the operation of step S1 is continued until the candidate of the target object, that is, the landmark (landmark), is detected, and the vehicle camera 11 continues to monitor the front of the host vehicle. Will do.

  On the other hand, when the image processing unit 14 detects a vehicle that can be a candidate for an object, that is, a landmark (landmark), in the traveling direction of the host vehicle (in the case of “target candidate detection” in step S1), the vehicle travels. In order to determine whether the vehicle is parked instead of being in the vehicle, the relative speed with respect to the host vehicle is detected based on the sensor information from the sensor 12 (step S2).

  The object position relation calculation unit 15 calculates the sum of the speed of the host vehicle and the detected relative speed, and whether the detected target vehicle is stopped or traveled depending on whether the sum is “0” or not. It is determined whether or not it is being performed (step S3). If the calculated sum is not “0” (in the case of “not 0” in step S3), the vehicle detected as a candidate for the target object, that is, a landmark (landmark) object, has not stopped (parking). It is determined that the vehicle is not a vehicle) and the process returns to step S1 to continue monitoring the front of the vehicle again.

  On the other hand, when the calculated sum is “0” (in the case of “0” in step S3), the vehicle detected as a candidate for the object, that is, the object that becomes a landmark, is a parked vehicle. And the outline of the target parked vehicle is read (step S4).

  To extract the contour of the parked vehicle, which is an object, that is, a mark (landmark) object, and detect the parked vehicle region, and then to determine the color of the parked vehicle, that is, the parked vehicle Then, a color name table CNT (Color Name Table) preset as a database is read (step S5). The color name table CNT is a compatibility table that holds a correspondence relationship between a detected color code (for example, a combination of RGB) and a color name used when outputting language information or graphic / character information. By replacing the color name table CNT according to the driving scene (for example, switching between daytime and nighttime) and the characteristics of the driver (personality related to color vision), for example, even if the color is the same red, “reddish color '', `` Crimson '', `` sunset color '', `` flame red '', `` bronze color '', `` red '', etc., depending on the scene and driver characteristics, the object, It becomes possible to control the conversion method into the language information and graphic / character information used when the information (features) related to the object to be the landmark (landmark) is presented to the driver.

  Next, the color name held in the color name table CNT as color information is detected for each unit plane (unit cell) of the area of the target parked vehicle read as image information (step S6). After that, the color name detected for each unit plane (unit cell) in the area of the target parked vehicle is information that characterizes the object whose color information is the object, that is, the landmark (landmark), as information that is easily visible to the driver. It is determined whether or not the color of the target parked vehicle is approximately by determining whether or not it occupies a ratio equal to or greater than a predetermined threshold for determining whether or not there is (step S7).

  As a color information detected for each unit plane (unit cell), when a specific color name occupies more than the threshold value, for example, 30% or more cells (Yes in step S7), the target parked vehicle When providing information on the driver to the driver, it is determined that it can be provided using the color name, and color information indicating 30% or more can be used as color guide information. (Step S8). At this time, when color names indicating a plurality of colors (for example, red and blue) are present at a ratio of not less than the threshold value, for example, 30% or more, a plurality of color names such as “red and blue” are selected. Each color name is retained in the format shown.

  On the other hand, as color information detected for each unit plane (unit cell), if any color name does not occupy more than the threshold, for example, 30% or more cells (No in step S7), The color information is not information that characterizes the target parked vehicle, but is not information that is easily visible to the driver, and is not held as guide content, and the process proceeds to step S9. Note that the threshold value determined in advance for determining whether or not the color information is information that characterizes an object that is an object, that is, a landmark (landmark), can be appropriately changed and set to an arbitrary ratio.

  Next, the process proceeds to processing for determining the shape and size of the target parked vehicle. First, preparations are made for reading the license plate of the subject parked vehicle. That is, it is confirmed whether or not the license plate can be read (step S9). For example, when the target parked vehicle is parked sideways (in the case of No in step S9), the license plate cannot be read, and the process proceeds to step S13.

  On the other hand, if the vehicle is parked in a state where the license plate is visible and the license plate can be read (Yes in step S9), the position of the license plate is detected and preparation for reading the license plate is performed (step S9). S10). Next, information of a classification number (for example, “220” in the case of “Sagami 220-xxx”) is read from information written on the license plate (step S11). Furthermore, using the read classification number, the vehicle shape table that is set in advance as a database, that is, the classification number table CLT (Classified Number Table), is used to determine what the classification (vehicle type) of the target parked vehicle is. Then, the vehicle shape of the vehicle type corresponding to the classification number is extracted (step S12).

  Next, the length of the target parked vehicle in the vertical and horizontal directions is estimated from the size of the contour of the target parked vehicle detected in step S4 (step S13). Next, using the lengths in the vertical and horizontal directions of the target parked vehicle, the vehicle size table VST (Vehicle Size Table) set and registered in advance as a database is used to indicate the outline of the target parked vehicle in the vertical and horizontal directions. If the length of the vehicle falls within one of the pre-registered areas, the vertical / horizontal length of the target parked vehicle is Information (features) related to an object that is a landmark is converted into language information, graphic / character information used when presenting to the driver (step S14). For example, when it is estimated that the target parked vehicle is a vehicle having a width (horizontal) of 2,300 mm or more and a height (vertical) of 2,800 mm or more, it is converted into an expression method of “large vehicle”.

  On the other hand, when the length of the target parked vehicle in the vertical and horizontal directions does not fit in any of the pre-registered areas, that is, it corresponds to any area of the vehicle size table VST set and registered in advance. If not, the length or size in the vertical / horizontal direction is determined not to be a characteristic characterizing the target parked vehicle, and is not held as guide information.

  Here, in the present embodiment, as information for characterizing the vehicle shape / size of the target parked vehicle, information on the vehicle shape based on the classification number obtained from the license plate of the target parked vehicle in step S12, or step In S14, any one of the information regarding the size of the vehicle based on the length in the vertical and horizontal directions obtained from the contour information of the target parked vehicle is used. For this reason, it is discriminate | determined whether the classification number was acquired from the number plate of the object parked vehicle in step S11 (step S15).

  If the classification number has been acquired (Yes in step S15), in step S12, information relating to the vehicle shape of the target parked vehicle obtained based on the classification number is provided to information that characterizes the target parked vehicle, that is, the driver. Is stored as possible guide information (step S16). On the other hand, if the classification number has not been acquired (Yes in step S5), the size of the vehicle obtained based on the length in the vertical and horizontal directions obtained from the contour information of the target parked vehicle in step S14. Is stored as information characterizing the target parked vehicle, that is, as guide information that can be provided to the driver (step S17).

  Next, it is confirmed whether or not there is signboard information in which characters / symbol mark information and the like are written on the body or roof of the target parked vehicle (step S18). When some character information is described on the signboard of the target parked vehicle or the roof (when “character information is present” in step S18), the character information described on the signboard of the target parked vehicle body or roof Is read (step S19). Next, it is searched whether or not the unique expression is included in the expression of the text composed of the read character information (step S20).

  Here, the proper expression is made up of proper nouns or combinations thereof, and it is determined how far a word string composed of a plurality of words is an individual name or a company name as proper nouns or combinations thereof. Note that a morphological analysis dictionary MAD (Morphological Analysis Dictionary) set in advance as a database is used in order to perform morphological analysis of the character string read in detecting the specific expression.

  When the read character information includes a specific expression (when “the specific expression exists” in step S20), the included specific expression may be provided to information that characterizes the target parked vehicle, that is, the driver. It is stored as signboard information (character information) indicating possible guide information (step S21). On the other hand, if the read character information does not include a unique expression (“no specific expression” in step S20), the signboard information (character information) is not information (characteristic) characterizing the target parked vehicle. judge.

  In addition, if any symbol mark is written on the signboard of the body or roof of the target parked vehicle in step S18 (when “symbol mark information is present” in step S18), the signboard of the body or roof of the target parked vehicle. Are read (step S22). Next, in order to determine whether or not the read symbol mark is a known symbol mark that can be provided as guide information, a symbol mark database SMD (Symbol Mark Database) set in advance as a database is searched (step S1). S23).

  When a match with the symbol mark recorded in the symbol mark database SMD is obtained (when “symbol mark information exists” in step S23), the symbol mark database SMD is recorded corresponding to the matched symbol mark information. The voice information and graphic / character information that is provided is held as signboard information (symbol mark information) indicating information characterizing the target parked vehicle, that is, guide information that can be provided to the driver (step S24). On the other hand, when the read symbol mark does not match any symbol mark in the symbol mark database SMD (in the case of “no symbol mark information” in step S23), the signboard information (symbol mark information) is the target parked vehicle. It is determined that it is not information (characteristic) that characterizes.

  By performing the above processing, when the image processing unit 14 serving as a mark extraction unit extracts another parked vehicle as the mark object, the color, shape, size, or parked vehicle of the parked vehicle is extracted. At least one of the signboard information drawn on the body of the car or laid on the roof is extracted as a characteristic characterizing the parked vehicle, but the extraction of the color, shape / size, and signboard information of the parked vehicle is over After that, each characteristic that characterizes the obtained parked vehicle, that is, each guide information is integrated, and the integrated characteristic, that is, the guide information is generated as presentation information regarding the object as a landmark and presented to the driver. In particular, as a feature characterizing the parked vehicle of the object as a landmark, any one of the color, shape, size, or signboard information drawn on the body of the parked vehicle or the facility or laid on the roof When a feature to be used as guide information is obtained, guide information in which a plurality of features are integrated is generated, and the guide information in which a plurality of features are integrated is omitted according to the route guide generation rule illustrated in FIG. Etc., and optimize as route guide information. FIG. 3 is a table showing an example of a route guide generation rule applied to the navigation device according to the present invention.

As shown in FIG. 3, as the integrated guide information when the guide information indicating the characteristics of the target parked vehicle is obtained,
(1) Presence or absence of vehicle color information,
(2) Presence / absence of vehicle shape / size information (3) There are 7 combinations of presence / absence of vehicle signage information except for the case of “None”, and only one of them can be obtained 3 types of pattern 1, one type of pattern 2 when color information and shape / size information are obtained, and one type of pattern 3 when other information including signboard information is obtained at the same time It is classified into ~ 5.

  As shown in the “Route Guide” column of FIG. 3, for example, when only the color information of the vehicle is obtained as the integrated guide information of the pattern 1, the guide information “the intersection where the red car stops” is output. If only the vehicle shape / size information is obtained as the integrated guide information of pattern 1, the guide information “the intersection where the bus stops” is output, and the integrated guide of pattern 1 is also output. When only the signboard information of the vehicle is obtained as information, the guide information “The intersection where the Odakyu bus stops” is output.

  When the color information and the shape / size information are obtained as the integrated guide information of the pattern 2, the guide information “the intersection where the red bus stops” is output, and the integrated guide information of the pattern 3 is output. When the color information and the signboard information are obtained, the guide information “The intersection where the red Odakyu bus stops” is output.

  On the other hand, when signboard information consisting of shape / size information and character information is obtained as the integrated guide information of pattern 4, guide information “the intersection where the Odakyu bus stops” is output, and the pattern information When the sign information including the color information, the shape / size information, and the character information is obtained as the integrated guide information 5, guide information “The intersection where the red Odakyu bus stops” is output. In the patterns 4 and 5, the same word “bus” is output redundantly in this state.

  In addition, the case where the symbol mark is acquired as the signboard information is the same as the case of the character information, which is not shown in the figure. For example, “the intersection where the red bus with the red Odakyu bus mark stops "...", the same word may be output twice.

  In other words, the possibility of redundant expression as guide information is limited to the case where signboard information can be acquired at the same time as other information as shown by the broken line in FIG. Only in the case of patterns 3 to 5 surrounded by a one-dot chain line, that is, when signboard information is acquired simultaneously with other information, whether or not there is a redundant expression in the expression contents, that is, the redundancy of the generated route guide information Check.

  Therefore, when the signboard information can be acquired simultaneously with other information, it is checked whether or not the redundant information is included in the integrated guide information, and the redundant expression is deleted. Optimize as information. Thereafter, the optimized guide information is output as linguistic information, graphic / character information, and the operation as navigation is terminated.

  In the flowchart of FIG. 2, as an example of the case of a parked vehicle parked at a guide point as an object of guide information added as route guide information, that is, an object to be a landmark (mark), Even when a building (building or signboard) installed on the roadside of a point set in advance as a point to be guided, that is, a facility is used as an object to be a landmark (mark), it is exactly the same as the flowchart of FIG. In addition, it can be generated as guide information and presented to the driver.

  For example, as a facility that can be used as a landmark, it emits light like a neon sign, is painted with a conspicuous color, is huge in size such as a skyscraper, or like a tower Any facility can be used as long as it has a feature that is easy to see for the driver, such as a conspicuous shape, a large sign on the written signboard, or a symbol mark with a conspicuous shape. Absent.

  However, when extracting the shape of the facility as a feature, the following processing is performed. In the description of the flowchart of FIG. 2, when extracting the shape of a parked vehicle as a characteristic characterizing the parked vehicle, a result of referring to a vehicle shape table of a vehicle type that is set in advance based on a result of reading a license plate, that is, a classification number table CLT On the other hand, if the shape of the corresponding vehicle is extracted and the license plate of the vehicle cannot be read, it is determined that the shape is not a characteristic characterizing the parked vehicle.

  On the other hand, when a facility shape is extracted as a feature characterizing the facility, a facility shape table registered in advance is provided, and instead of referring to the classification number table CLT, it becomes a landmark (a landmark). Based on the pattern matching result between the facility shape detected as an object and the facility shape table, the shape of the corresponding facility is extracted. On the other hand, there is a pattern that matches the shape of the facility in the facility shape table. If not, it may be determined that the shape is not a feature that characterizes the facility.

  As described in detail above, according to the navigation device and the navigation method of the present embodiment, the following effects can be obtained.

  At the point to be guided, an object to be a landmark (landmark) is actually extracted from detected objects (for example, an object photographed by the vehicle-mounted camera 11) and presented to the driver as guide information. It is guaranteed that the object (landmark), that is, the object actually exists, and the object (other vehicle or roadside facility) that is easy to see from the driver is guided to the driver. Thus, it is possible to accurately and easily grasp an object that becomes a landmark (landmark), that is, an object.

  In other words, unlike the conventional technology, it is not old information from the map database that cannot cope with newly added facility information or deleted facility information after the creation time, but it can be actually seen by the driver It is possible to present, as guide information, audio information and / or character / graphic information indicating the characteristics of a captured image of an object that currently exists as a landmark (landmark) and, in some cases, the distance to the object.

  Therefore, in the guide information using the information recorded in the map database in the prior art, the situation that is inconsistent with the current situation due to the construction near the actual intersection (the facility recorded in the map database disappears or another facility It was not possible to cope with the occurrence of such a situation that it was replaced or that a new facility was built in the vicinity. However, in the present embodiment, since the guide information is based on information actually read by the in-vehicle camera 11, there is actually an object to be presented as guide information, that is, an object serving as a landmark (landmark) at the guide point. It is guaranteed that

  In addition, in the case of a map database, depending on the shape of the intersection to be guided, for example, the shape of the intersection to bend, even if the facility is invisible near the driver, it is only present in the vicinity of the intersection. In other words, an object that becomes a landmark (landmark) may be used in the guide information. However, in the case of this embodiment, such a situation does not occur, and only an object that can be surely seen by the driver is used for the guide information. .

  In addition, even if the intersection and the landmark facility are displayed in the vicinity on a flat map in a general map database, even if the landmark is visible from the direction of the driver, the intersection and the landmark are displayed. It does not always appear in an easily understandable position. Alternatively, there is no guarantee that facilities that should be landmarks can be confirmed from the driver's direction.

  In addition, when facility information recorded on the map database is used as guide information, information that can be used as a guide such as signboard information in which the facility information is described is not installed toward the driver side, guide Even if presented as information, the driver cannot confirm the guide information. On the other hand, in the case of the present embodiment, since it is guide information based on an object actually read by the in-vehicle camera 11, that is, an object to be a landmark (landmark), information that cannot be seen by the driver is not used as guide information. What is easy for the driver to visually recognize is presented as guide information.

  Furthermore, the features that characterize other vehicles and facilities that can be read landmarks can be presented to the driver as language information and / or text / graphics information that is easy for the driver to understand, improving the ease of understanding guide information. be able to.

(Second Embodiment)
Next, as a second embodiment of the navigation device and the navigation method according to the present invention, a selection method when a plurality of target parked vehicles exist near the intersection of the guide points will be described with reference to the flowchart of FIG. . FIG. 4 is a flowchart showing an example of the navigation apparatus according to the present invention which is different from the operation of FIG. 2, and shows another example of the navigation method according to the present invention. Note that the block configuration of the navigation device in the present embodiment may be the same as that of the navigation device shown in FIG. 1 as the first embodiment.

  The flowchart of FIG. 4 is the same as the flowchart of FIG. 2, when the vehicle approaches a predetermined point as a point to be guided during route guidance, that is, reaches the guide timing at which guide information is to be output. When activated, it is activated to present guide information to the driver as presentation information.

  In the flowchart of FIG. 4, similarly to step S <b> 1 of FIG. 2, first, a vehicle that can be a candidate for an object, i.e., an object to be a landmark (landmark), is guided in the traveling direction of the host vehicle using the on-vehicle camera 11. It continues to monitor whether or not it exists in the vicinity of the intersection of the points (step S31). When the image processing unit 14 refers to the image information captured by the in-vehicle camera 11 and stored in the memory, when a vehicle that should be a candidate for an object, that is, a landmark (landmark), is not detected (step S31). In the case of “No”, the operation of step S31 is continued until the candidate of the object, that is, the landmark (landmark) object is detected, and the vehicle camera 11 continues to monitor the front of the host vehicle. Become.

  On the other hand, when the image processing unit 14 detects a vehicle that can be a candidate for an object, that is, a landmark (landmark) object, in the traveling direction of the host vehicle (in the case of “Yes” in step S31), FIG. As described in the flowchart, it is confirmed whether or not the vehicle is a parked vehicle, and if it is a parked vehicle, as the landmark candidate, the color information of the vehicle, the vehicle shape / size information, the signboard information, etc. The features of the target parked vehicle candidate are extracted (step S32). Thereafter, as long as another vehicle is detected in the traveling direction of the host vehicle (in the case of “Yes” in Step S33), the feature extraction process of the target parked vehicle candidate in Step S32 is repeated, and a plurality of landmark candidates, The characteristics of each of the plurality of target parked vehicle candidates are extracted.

  When the feature extraction processing for all landmark candidates existing in the vicinity of the intersection in the traveling direction of the host vehicle, that is, the target parking vehicle candidates has been completed (in the case of “No” in step S33), then a plurality of target parkings are performed. Each vehicle candidate identifies which zone (ZONE) in the vicinity of the intersection of the guide points is parked.

Here, as a zone (ZONE) to be identified, as shown in FIG. 5, a zone where the vehicle around the relevant intersection and the vehicle in the intersection may park and stop is viewed from the own vehicle,
(1) Zone around the intersection on the road where the vehicle is traveling (ZONE1)
(2) Zone around the intersection on the road where the vehicle will turn left and right according to the route guide information obtained as a navigation function to the destination (ZONE2)
(3) Stop zone (ZONE3) of the vehicle in the intersection when the host vehicle turns right or left
It classifies into three. FIG. 5 is a schematic diagram showing a state in which the area around the intersection where the vehicle may park and stop and the area in the intersection are divided into zones.

Furthermore, about the zone (ZONE1) around the intersection on the road where the vehicle is traveling, as shown in FIG.
(1-1) Zone before the intersection on the traveling route of the host vehicle (ZONE1-1)
(1-2) Zone beyond the intersection on the opposite line of the vehicle (ZONE1-2)
(1-3) Zone on the other side of the intersection on the traveling route of the own vehicle (ZONE1-3)
(1-4) Zone before the intersection on the opposite line of the vehicle (ZONE1-4)
These are subdivided into four zones, and together with ZONE2 and 3, for convenience, it is classified into six zones (ZONE).

As described above, as shown in FIG. 4, in the order in which it is estimated that the recognition of the target parked vehicle is easy when viewed from the driver, for each zone classified into the vicinity of the intersection of the guide point and the inside of the intersection into six, For example, in order from ZONE1-1,
ZONE1-1>ZONE1-2>ZONE3> ZONE2
>ZONE1-3> ZONE1-4
The priority is given in the order of, and that fact is registered in the priority table.

  Therefore, in the flowchart of FIG. 4, first, the priority table is read (step S34), and a plurality of target parked vehicle candidates are identified in which zone (ZONE) around the guide point intersection. The target parking vehicle candidate parked in the highest priority zone (ZONE), for example, ZONE 1-1 among the target parked vehicle candidates parked or parked, a point to be guided, for example, an intersection guide to bend The vehicle is selected as a target parking / stopping vehicle (step S35).

  Regarding the target parked vehicles selected as landmarks from multiple target parked vehicle candidates, the features such as vehicle color information, vehicle shape / size information, signboard information, etc. previously extracted The information integrated and optimized as described in the first embodiment is added to the route guide information obtained as the navigation to the destination as the guide information at the intersection of the points to be guided, and the driver To present.

  Note that the priority for each region (zone) registered in the priority table can be changed from the priority shown in FIG. 4, but information on visibility from the own vehicle, and It is important to give priority to the information related to the direction of travel so that the most easily visible object for the driver can be selected as an object that will be a landmark (marker), and it will be important to give consideration to improving the understanding. It is.

  As described above, in the present embodiment, among a plurality of target parked vehicle candidates, according to the priority regarding the parked area (zone), that is, for each area (zone) of a predetermined point in advance. According to the specified priority order, the target parked vehicle candidate parked in the highest priority area is selected as an object to be a landmark, that is, the target parked vehicle, and the selected target parked vehicle In addition to the effects of the first embodiment, the vehicle is presented to the driver using the voice information and / or character / graphic information as guide information by the information generation / output unit 30 of the information generation / output means. It is possible to provide guide information that is easier for the driver to visually recognize.

  4 and 5, as in the case of the first embodiment, the case where the object, that is, the object to be a landmark (mark), is a parked vehicle has been described. However, the landmark (mark) is used. An object is not limited to a parked vehicle, but any object that serves as a mark indicating the characteristics of a point to be guided, such as a building installed at a point to be guided, that is, a roadside facility. It doesn't matter.

It is a block block diagram which shows the block configuration of 1st Embodiment of the navigation apparatus by this invention. It is a flowchart which shows an example of operation | movement of the navigation apparatus shown in FIG. It is a table which shows an example of the route guide production | generation rule applied to the navigation apparatus by this invention. 3 is a flowchart showing an example different from the operation of FIG. 2 of the navigation device according to the present invention. It is a schematic diagram which shows a mode that the area | region around the intersection in which a vehicle may park and stop, and the area | region in an intersection are divided into zones.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Object recognition part, 11 ... Camera (vehicle-mounted camera), 12 ... Sensor (vehicle-mounted sensor), 13 ... Image holding part, 14 ... Image processing part, 15 ... Object positional relationship calculation part, 20 ... Route guide calculating part , 21 ... route information comparison calculation unit, 22 ... route calculation unit, 23 ... map database, 30 ... information generation / output unit, 31 ... information generation unit, 32 ... speaker, 33 ... display unit.

Claims (19)

  A navigation device that is mounted on a vehicle, calculates a route to a destination, and guides the vehicle to the calculated route, and includes an object detection unit that detects an object existing around the vehicle at a point to be guided, and the object detection unit It comprises at least mark extraction means for extracting an object as a mark from the detected objects, and information generation / output means for generating and outputting presentation information to the driver corresponding to the object extracted as the mark. A featured navigation device.   2. The navigation device according to claim 1, wherein the information generation / output means outputs the presentation information when a vehicle approaches a predetermined point as a point to be guided during route guidance. A navigation device.   3. The navigation device according to claim 1, wherein the information generation / output unit outputs the presentation information as audio information and / or outputs as character / graphic information.   4. The navigation device according to claim 1, wherein the object detection unit includes a camera that captures an object around the vehicle and a sensor that measures a distance between the vehicle and the captured object. The navigation apparatus characterized by being made.   5. The navigation device according to claim 1, wherein the landmark extraction unit is a vehicle that is parked or stopped around a predetermined point from the objects detected by the object detection unit, or a predetermined unit. A navigation apparatus characterized in that facilities existing around a certain point are extracted as objects as landmarks.   6. The navigation device according to claim 1, wherein the landmark extracting unit detects a plurality of candidate objects detected as landmarks when a plurality of candidate objects are detected around a predetermined point. Are selected as objects that are candidates for objects existing in areas with higher priorities according to the priority order determined for each of the predetermined point areas, and the information generation / output means A navigation device characterized in that the presentation information relating to the selected object is generated and output.   The navigation device according to claim 5 or 6, wherein when the other vehicle or facility is extracted as an object used as the landmark, the landmark extracting unit extracts the color, shape, size, At least one of signboard information drawn on a vehicle body or the facility or signboard information laid on the roof of the vehicle or the facility is extracted as a feature characterizing the vehicle or the facility Navigation device.   The navigation device according to claim 5 or 6, wherein the mark extracting means is characterized by a color, shape, size, vehicle body of the vehicle, or a sign drawn on the facility as a characteristic characterizing the vehicle or facility of the object to be the mark. If any of a plurality of features is extracted from the information or signboard information laid on the roof of the vehicle or the facility, the extracted features are integrated and integrated by the information generation / output means A navigation device characterized in that the generated feature is generated and output as the presentation information relating to an object as a landmark.   9. The navigation device according to claim 7 or 8, wherein the landmark extracting means extracts the color of another vehicle or facility as a characteristic characterizing the vehicle or the facility, for each unit plane of the vehicle or the facility. The color code of the color that occupies a ratio equal to or greater than a predetermined threshold is extracted as a feature. On the other hand, if there is no color that occupies a ratio equal to or greater than the threshold, the color is the vehicle or A navigation apparatus characterized by determining that the facility is not a characteristic characterizing the facility.   10. The navigation device according to claim 9, wherein the information generating / outputting unit uses a color code extracted as a characteristic characterizing another vehicle or facility by the mark extracting unit, and indicates a preset color code. A navigation apparatus characterized by converting information and / or character / graphic information and outputting the information.   The navigation device according to any one of claims 7 to 10, wherein the mark extracting means extracts in advance the shape of another vehicle or facility as a characteristic characterizing the vehicle or the facility based on a result of reading a license plate of the vehicle. Based on the result of referring to the vehicle shape table of the set vehicle type, or based on the pattern matching result between the shape of the facility and the facility shape table registered in advance, the corresponding shape is extracted as a feature, On the other hand, if the license plate of the vehicle cannot be read, or if there is no pattern that matches the shape of the facility in the facility shape table, the shape of the vehicle or the facility is A navigation device characterized by determining that it is not a feature to be characterized.   12. The navigation device according to claim 11, wherein the information generating / outputting unit includes a linguistic information indicating the preset shape, the shape extracted as a characteristic characterizing another vehicle or facility by the landmark extracting unit, and A navigation device characterized by converting to character / graphic information and outputting it.   The navigation device according to any one of claims 7 to 12, wherein the landmark extracting means extracts the size of another vehicle or facility as a characteristic characterizing the vehicle or the facility. When the length in the vertical and horizontal directions indicating the contour is within one of the pre-registered areas, the length in the vertical and horizontal directions of the vehicle or the facility is extracted as a feature, while the vehicle or If the vertical / horizontal length of the facility does not fit in any of the pre-registered areas, it is determined that the vertical / horizontal length is not a characteristic characterizing the vehicle or the facility. A navigation device characterized by that.   14. The navigation device according to claim 13, wherein the information generating / outputting unit is a language indicating the size set in advance as a size extracted as a characteristic characterizing another vehicle or facility by the landmark extracting unit. A navigation apparatus characterized by converting information and / or character / graphic information and outputting the information.   The navigation device according to any one of claims 7 to 14, wherein the landmark extracting means extracts signboard information of another vehicle or facility as a characteristic characterizing the vehicle or the facility. When character information is present in the signboard information and the character information includes a character string consisting of a proper noun or a combination thereof, the proper noun or a combination thereof is extracted as a feature, A navigation device characterized in that when the character information does not include a character string composed of proper nouns or a combination thereof, the character information of the signboard information is determined not to be a feature characterizing the vehicle or the facility.   16. The navigation device according to claim 15, wherein the information generating / outputting unit includes a proper noun or a combination of the proper noun extracted as a characteristic characterizing another vehicle or facility by the landmark extracting unit. Alternatively, a navigation device characterized in that it is converted into linguistic information and / or character / graphic information indicating the combination.   The navigation device according to any one of claims 7 to 16, wherein the landmark extracting means extracts signboard information of another vehicle or facility as a characteristic characterizing the vehicle or the facility. If a symbol mark is present in the signboard information and the symbol mark is the same as one of the preset symbol marks, the symbol mark is extracted as a feature, When the symbol mark does not match any of the preset symbol marks, the navigation device is characterized in that the symbol mark of the signboard information is determined not to be a feature characterizing the vehicle or the facility. .   18. The navigation device according to claim 17, wherein the information generating / outputting unit includes a symbol mark extracted as a characteristic characterizing another vehicle or facility by the landmark extracting unit, and a language indicating the preset symbol mark. A navigation apparatus characterized by converting information and / or character / graphic information and outputting the information.   A navigation method that is mounted on a vehicle, calculates a route to a destination, and guides the vehicle to the calculated route, detects an object existing around the vehicle at a point to be guided, and serves as a mark from the detected objects A navigation method, wherein an object is extracted, and presentation information to a driver corresponding to the extracted object is generated and output.

Images