JP2006038557A - Car navigation system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily input character information indicated in an object such as a magazine etc. as locational information for specifying a destination and registering memory locations and improve reliability in specifying the destination at the time. <P>SOLUTION: This car navigation system 1 is provided with a method for inputting locational information such as facilities and stores which a user desires to specify, addresses, and telephone numbers on the basis of character recognition from images inputted by the user as one of methods for specifying a destination (or for registering memory locations) by the user. In this case, the user inputs an image of a document such as a magazine containing the locational information by photographing it with an onboard camera 2. A control circuit 5 reads image data from the onboard camera 2, displays it on a screen of a display part 7, and makes an image recognition part 10 recognize character information. The locational information is extracted from the recognized character information to execute processing for specifying destination candidates. In the case that a plurality of types of locational information are extracted, they are combined to narrow down destination candidates. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a car navigation device that allows a user to input point information for destination designation and memory point registration.

  A car navigation device mounted on an automobile includes a main body including a microcomputer, a display device, a GPS receiver, a map database (a medium such as a DVD storing road map data and its drive device), and the like. It also has a location function that detects its current position and displays it on the display device in a superimposed manner on a road map, and a route guidance function that searches and guides a recommended route to the destination specified by the user. Yes.

  At this time, the map database stores destination data (index data consisting of name, address, telephone number, etc.) of various facilities and stores in association with road map data. In addition, the user can register a frequently used point as a destination (for example, a home or an acquaintance's house) as a memory point. In order to execute the route guidance function, when the user designates the destination, by inputting the name and address of the facility by the operation of the user's touch panel or the like, the corresponding facility is determined from the destination data. You can search and identify the location.

  By the way, a user may want to go to a facility or store introduced in a magazine, for example. In this case, the user can bring the magazine into the car and enter the name and address while looking at it. Can be designated as the ground. However, when the name is directly specified, the user needs to input the hiragana character by character, and when specifying the phone number or address, the user can input a number character by character or select a list. There is a problem that it takes a lot of time and effort for the user operation.

In this case, a technique for acquiring address information and the like by capturing image data from a document using a scanner and performing character recognition (OCR) from the image data has been conventionally known (see, for example, Patent Document 1). However, it is not easy to bring image data from a document by bringing a scanner into the car.
JP 2001-157876 A

  As described above, if information such as names and addresses of facilities and shops published in magazines can be easily input to the car navigation device as character information (point information), it is possible to reduce the labor of the user. It can be easy to use. Therefore, as a portable portable image input device, for example, a camera-equipped mobile phone, which has recently become widespread, is used to photograph magazine publication information, and the image data is transferred to a car navigation device where character recognition is performed. It is also possible to make it happen. However, because the address, etc. published in a magazine is often incomplete, such as omitting the prefecture name, city name or street address due to space and description format, it is inferior in certainty of specifying the destination. Therefore, it is necessary to devise in order to increase the certainty of the destination.

  The present invention has been made in view of the above circumstances, and its purpose is to enable easy input of character information written on an object as point information for destination designation and memory point registration. The present invention provides a car navigation device that can increase the certainty of a destination at that time.

  In order to achieve the above object, the car navigation device of the present invention searches a map database based on point information input by a user for destination specification or memory point registration, thereby finding a corresponding destination. An image data input means for inputting an image of an object on which the point information is described, and a recognition means for recognizing character information included in the image and extracting it as point information And a search means for searching the map database based on the spot information extracted by the recognition means to select destination candidates, and the search means includes a plurality of types of points by the recognition means. When information is extracted, it is characterized in that it is configured to narrow down destination candidates by combining these multiple types of point information ( Invention of Motomeko 1).

  According to this, the user can input an image of an object (for example, a document such as a magazine) in which point information is described by the image data input means, and the character included in the input image by the recognition means. Information is recognized and extracted as point information. Therefore, it is possible to easily input point information such as names, addresses, and telephone numbers of facilities and stores published in objects such as magazines as character information, and it is easier to use by reducing the user's trouble. can do.

  Based on the spot information extracted by the recognition means, the search means searches the map database and selects destination candidates. The spot information extracted by the recognition means is, for example, one There is a possibility that the portion is incomplete, and in this case, it is considered that the destination candidates cannot be narrowed down to one. However, since the search means is configured to be able to narrow down the destination candidates by combining a plurality of types of point information extracted by the recognition means, it is possible to select a destination with higher probability. As a result, the certainty of the destination can be increased.

  By the way, in recent years, in vehicles such as automobiles, it is considered to provide an in-vehicle camera for photographing the front and surroundings of the vehicle. Therefore, in the above configuration, the in-vehicle camera provided in the vehicle can be used as the image data input means (the invention of claim 2), thereby eliminating the need to separately provide the image data input means and simplifying the configuration. I can finish it.

  More specifically, the recognizing means is configured to extract at least one or more types of data out of a point code obtained by coding an address, a telephone number, a zip code, and a position as point information. (Invention of claim 3), according to this, the certainty of specifying the destination by the search means can be enhanced by the combination of the plurality of types of point information. Further, at this time, when the character information extracted as the point information includes kanji, it is determined that the search means is address data, and when the character information is composed of numbers, a telephone number or a zip code Therefore, it is possible to make a search by judging that the data is any one of the point codes (invention of claim 4), whereby the search can be performed efficiently.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. First, FIG. 1 schematically shows the overall configuration of a car navigation apparatus 1 according to this embodiment incorporated in a vehicle (automobile). In this embodiment, the vehicle is mounted on a vehicle as image data input means. The vehicle-mounted camera 2 can be used. Here, the car navigation device 1 includes an own vehicle position detector 3, an image input unit 4, a control circuit 5 connected thereto, a map database 6 connected to the control circuit 5, a display unit 7, an input device unit. 8, an audio output device 9, an image recognition unit 10 and the like.

  Among them, the own vehicle position detector 3 is a GPS receiver 11 for GPS (Global Positioning System) that detects (positions) the position of the own vehicle based on a radio wave transmitted from an artificial satellite for GPS, and the rotational angular velocity of the own vehicle. Has a gyroscope 12, a vehicle speed sensor 13 for detecting the traveling speed of the host vehicle, and a geomagnetic sensor 14 having a known configuration. Since the sensors 11 to 14 of the position detector 3 have errors having different properties, the control circuit 5 is configured to be used while being interpolated. The current position, traveling direction, speed, travel distance, current time, etc. of the vehicle are detected with high accuracy. Depending on the accuracy, a part of the above-described sensors 11 to 14 may be omitted, and in addition to the above, a steering rotation sensor, a distance sensor, or the like may be employed.

  The map database 6 includes a medium storing various data such as road map data and destination data (index data) accompanying the road map data, and a drive device for inputting data from the medium. -A mass storage medium such as a ROM, DVD, or hard disk is used. The road map data includes data such as road shapes, road names, buildings, various facilities, their telephone numbers, place names, and terrain, and the road map is reproduced on the screen of the display unit 7. It is comprised including the data of.

  The destination data includes information on facilities such as transportation facilities, leisure facilities, accommodation facilities, public facilities, and various stores such as retail stores, department stores, restaurants, etc., and displays landmarks indicating these facilities. It is configured to include data for displaying on the road map superimposed on the screen of the unit 7. At this time, this destination data is the location data of each facility, store, etc., the type, name, address, postal code, telephone number, map code (a location code in which the location is coded. It is configured so that it can be searched from. The user can register a frequently used destination (for example, a home or an acquaintance's house) as a memory point and store it in a memory (not shown).

  The display unit 7 is composed of, for example, a liquid crystal display, and a road map of various scales is displayed on the screen, and a pointer indicating the current position and the traveling direction of the vehicle is displayed so as to be superimposed on the display. It has become. In addition, various input screens for the user to input a destination, various messages, and the like are also displayed. Furthermore, when the route guidance function for guiding to the destination is executed, the route to be overlapped on the road map is displayed. At this time, as will be described later, when the destination is specified from the image data (or when the memory point is registered), the input image is displayed on the screen of the display unit 7 (FIG. 3). (See (a)).

  Although not illustrated in detail, the input device unit 8 includes a mechanical switch provided in the vicinity of the screen of the display unit 7 and a touch panel provided on the screen of the display unit 7. Various commands such as designation of a destination and selection of a scale of a road map displayed on the display unit 7 can be input using the input device unit 8. In addition, a microphone (voice input device) for a user to input a command by voice, a remote controller, and the like are also provided. The voice output device 9 is configured to output various necessary messages using synthesized voice from a speaker during route guidance or the like.

  The control circuit 5 is composed mainly of a known microcomputer comprising a CPU, ROM, RAM, etc., and controls the entire car navigation apparatus 1. At this time, the control circuit 5 realizes a location function for knowing the current position of the host vehicle by the software configuration, and recommends a route to a destination designated based on the map data of the map database 6. A route guidance function for searching for and guiding the user is realized.

  Among them, the location function displays the road map on the screen of the display unit 7 based on the map data from the map database 6 as described above, and also determines the current position and progress of the vehicle based on the detection of the vehicle position detector 3. The current location mark indicating the direction is displayed. In this case, as the vehicle travels, the display of the current position moves on the map, and the map is scrolled according to the position of the vehicle. At this time, map matching is performed in which the current position of the vehicle is placed on the road.

  In addition, the route guidance function allows the user to specify a recommended travel route from the starting point (current position) of the vehicle to the destination specified by the user based on the input operation of the input device unit 8 by the user. It is automatically calculated using a well-known Dijkstra method according to conditions (for example, toll road priority, general road priority, distance priority, recommendation, etc.) and, along with this, the driving route is displayed on the screen of the display unit 7 as a color different from other roads. Is displayed and guided to the destination, and further, for example, by outputting a guidance voice from the voice output device 9 such as “the intersection is 200 m ahead on the left” at a required timing.

  The image input unit 4 can be connected to a small device owned by the user, such as a camera-equipped mobile phone camera or a digital camera, and captures image data captured by the device into the car navigation device 1. Has been. The image recognition unit 10 is provided with an OCR function that processes image data input from the image input unit 4 or the in-vehicle camera 2 and recognizes it as character information.

  On the other hand, the in-vehicle camera 2 is composed of a CCD camera, for example, and is provided in the vehicle, for example. In this case, the in-vehicle camera 2 normally captures, for example, the front or the periphery of the vehicle, and the captured image data at that time is a display device (not shown) as, for example, ambient information for the driver. Or used for automatic detection of the inter-vehicle distance from the preceding vehicle. At this time, in the present embodiment, the in-vehicle camera 2 is provided such that the direction thereof can be arbitrarily changed (or the user can remove and freely move the image), and an image photographed by the in-vehicle camera 2 is provided. Data is configured to be input to the control circuit 5.

  As will be described later in the description of the operation, in the car navigation apparatus 1 according to the present embodiment, as one of the methods for specifying the destination (or registering the memory location) by the user, the image from the image input by the user is used. There is a method of inputting point information such as the name or address of a facility or store to be designated or a telephone number based on character recognition. In this case, the user directly captures an image of an object (for example, a document such as a magazine) on which the location information is posted, by using the above-described in-vehicle camera 2 or by using a camera or a digital camera of a mobile phone and inputs an image. 4 can be entered.

  In this method, the user first selects “image input” in a state where the initial screen for specifying the destination is displayed on the display unit 7, so that the control circuit 5 first has the vehicle-mounted camera 2 or the image input unit. The image data is taken in from 4 and displayed on the screen of the display unit 7 (see FIG. 3). Accordingly, the in-vehicle camera 2 and the image input unit 4 constitute image data input means.

  Next, the control circuit 5 causes the image recognition unit 10 to process the captured image data, recognizes character information included in the image, and extracts it as point information. At this time, the control circuit 5 extracts at least one type of point information as much as possible among the address, telephone number, postal code, and map code as the point information. Therefore, the image recognition unit 10 and the control circuit 5 function as recognition means. At this time, instead of processing the entire input image data, the user may designate a necessary portion (a range to be recognized as character information).

  Then, the control circuit 5 functions as a search unit that searches the map database (destination data) based on the recognized point information and selects candidate destinations to be specified. At this time, when a plurality of types of point information are extracted based on character recognition, the control circuit 5 combines the plurality of types of point information to narrow down destination candidates. It is designed to identify high destinations.

  Also, in this case, if the character information extracted as point information contains kanji, it is determined as address data. If the character information is composed of numbers, a phone number, postal code, point code A search is performed by determining that the data is any of the above. After the destination is selected (specified), the control circuit 5 performs route search and guidance (or registers as a memory point) based on the destination.

  Next, the operation of the above configuration will be described with reference to FIGS. Now, when the user wants to designate, for example, a store or the like published in a magazine as a destination for route guidance of the car navigation device 1 (or register as a memory location), the magazine is brought into the vehicle. Then, as described above, after selecting “image input” on the destination designation initial screen of the display unit 7, for example, by using the in-vehicle camera 2, a portion where the address of the corresponding store in the magazine is described. Take a picture. Then, the captured image data is taken into the car navigation device 1.

  The flowchart of FIG. 2 shows an outline of a processing procedure executed by the control circuit 5 until the destination is specified from the image data input from the in-vehicle camera 2. That is, first, in step S1, the captured image data is displayed on the screen of the display unit 7. In the next step S2, the image data is sent to the image recognition unit 10, and character recognition processing is performed. Then, in step S3, a process for extracting point information from the recognized character information and specifying a destination candidate is executed. At this time, when point information is extracted for a plurality of types, destination candidates are combined to narrow down destination candidates. A specific example of the processing will be described later. In step S4, the destination specified in step S3 is displayed on the display unit 7, taken in as navigation information based on the user's confirmation operation, etc., and route search (or memory point registration) processing is performed. It is executed.

  Next, a specific example of the processing in steps S2 and S3 in the case where the destination is specified by combining the plurality of types of point information described above will be described with reference to FIG. In this case, as shown in FIG. 3 (a), in the image (store location information published in a magazine or the like) photographed by the in-vehicle camera 2 and displayed on the display unit 7, the first line (1) A part of the address such as “Showacho 1-1” is displayed (prefecture and city names are omitted), and the second line (2) at the bottom of the address is “0566-61-****”. The case where the number is displayed is given as an example.

  The flowchart of FIG. 3B shows a procedure of processing executed by the control circuit 5 when the point information is extracted for a plurality of types (the above two types) and the candidates are combined to narrow down destination candidates. Yes. That is, first, in step S11, the character string in the first line (1) is recognized, and the map database is searched. In this case, character information such as “Showamachi 1-1” is recognized, but since this character information includes kanji, it is determined to be address data and a search is performed. In step S12, destination candidates are extracted. However, because this address data is incomplete, the results of the search are, for example, “Showa-cho, Minami-Akita-gun, Akita-ken”, “Showa-cho, Naka-koma-gun, Yamanashi-ken”, and “Showa-cho, Kariya-shi, Aichi”. Land candidates are searched.

  In the next step S13, the character string of the second line (2) is recognized, and the map database is searched. In this case, character information such as “0566-61-***” is recognized, but since this character information is composed of numbers, it can be a phone number, postal code, or point code. It is determined that there is a destination, and the destination candidate is searched. In step S14, destination candidates are extracted. Here, “Akari Prefecture Kariya City” is searched from the area code of the telephone number. In step S15, destination candidates are narrowed down (extracted with a high probability) from the above two search results (steps S12 and S14), and the destination is determined. In this example, “Showa Town, Kariya City, Aichi Prefecture” is selected as the destination.

  Thus, according to the present embodiment, the following effects can be obtained. That is, the user can input point information such as the name, address, and telephone number of the facility or store that the user wants to specify based on the character recognition from the image data captured by the in-vehicle camera 2. Therefore, unlike conventional ones, where the user has to enter hiragana and numbers one by one when specifying the name, information such as names and addresses of facilities and stores that have been published in objects such as magazines can be simplified. It can be input as character information (point information), and can be made easier to use by reducing the user's trouble.

  And even if some of the extracted point information is incomplete, it is configured to be able to narrow down the destination candidates by combining multiple types of extracted point information, so it is more probable A high destination can be selected, and as a result, the certainty of the destination can be increased. Further, in particular, in this embodiment, since the vehicle-mounted camera 2 is used as the image data input means, there is no need to separately provide the image data input means, and there is an advantage that the configuration can be simplified. be able to.

  In each of the embodiments described above, the in-vehicle camera 2 is used for inputting image data. However, the image may be input using other image data input means such as a digital camera or a camera-equipped mobile phone. . In the above-described embodiment, the case where address data and telephone number data are combined is given as an example of the type of point information. However, address data, telephone number data, postal code data, point data (map code data), etc. May be arbitrarily combined. At this time, a search with higher probability can be performed by combining three or more types of data. Further, it is possible to perform a search by changing the weight depending on the type of data, for example, increasing the weight for address data and telephone number data.

  In addition, the present invention is not limited to the above-described embodiment. For example, an object for inputting an image is not limited to a magazine, and may be a flyer of an advertisement, a business card, a catalog, a slip, In addition, various modifications can be made to the overall configuration of the car navigation apparatus, and the present invention can be implemented with appropriate modifications within a range not departing from the gist.

The block diagram which shows one Example of this invention and shows the whole structure of a car navigation apparatus roughly Flowchart showing an outline of a processing procedure executed by the control circuit until the destination is specified from the image data The specific example of a process in the case of specifying a destination by combining a plurality of types of point information is shown, (a) is a diagram showing a display example, (b) is a flowchart showing a processing procedure.

Explanation of symbols

In the drawings, 1 is a car navigation device, 2 is an in-vehicle camera (image data input means), 4 is an image input section (image data input means), 5 is a control circuit (search means), 7 is a display section, and 10 is image recognition. Part (recognition means).

Claims (4)

A car navigation device that identifies a corresponding destination by searching a map database based on point information input by a user for destination specification or memory point registration,
Image data input means for inputting an image of the object in which the point information is described;
Recognizing means for recognizing character information contained in the image and extracting it as point information;
Search means for searching the map database based on the point information extracted by the recognition means and selecting candidate destinations, and
The car navigation device characterized in that, when a plurality of types of spot information are extracted by the recognition means, the search means is configured to narrow down destination candidates by combining the plurality of types of spot information. .   2. The car navigation apparatus according to claim 1, wherein the image data input means is an in-vehicle camera provided in a vehicle.   The recognizing unit is configured to extract at least one kind of data from a point code obtained by coding an address, a telephone number, a zip code, and a position as the point information. Item 3. The car navigation device according to Item 1 or 2. The search means determines that the character information extracted as the point information is address data if the character information is included, and if the character information is composed of numbers, a telephone number, a zip code, 4. The car navigation apparatus according to claim 3, wherein the search is performed by determining that the data is any one of the point codes.

Images