JP5983299B2 - Feature information collection system, center, and feature information collection method - Google Patents

Description

  The present invention relates to a feature information collection system, a center, and a feature information collection method.

  Conventionally, a technique for registering a new feature newly established on a road has been developed. For example, Patent Document 1 discloses a technique in which a detection unit of a traveling vehicle recognizes a new feature, registers it in a database included in the vehicle, and uses it for next driving support. Moreover, in patent document 2, when the detection part of the own vehicle images the surrounding environment and the detected signboard is not registered in the database of the own vehicle, it recognizes that it is a newly installed signboard, and as a new feature A technique for adding a new signboard to a database is disclosed.

JP 2008-249479 A JP 2007-240198 A

  However, in the prior art, in order to specify a new feature detected by the detection unit of the own vehicle, a function for comparing with a large amount of candidate data is necessary, and this function is large when one vehicle is mounted. In addition, there is a problem that the processing load on the vehicle is high when executing this function. Thus, in the prior art, there is room for further improvement in terms of specifying new features while reducing the processing load on the vehicle.

  The present invention has been made in view of the above circumstances, and provides a feature information collection system, a center, and a feature information collection method capable of specifying a new feature while reducing the processing load on the vehicle. The purpose is to provide.

  The feature information collection system of the present invention is a feature in which a plurality of vehicles having at least a detection unit that detects a feature installed in a traveling direction of a road are associated with the feature and position information of the feature. And a center having at least a storage unit that stores feature parameters corresponding to a plurality of feature points that specify the feature, and the plurality of vehicles and the center are communicably connected. In the information collection system, when there are new feature candidates newly established on the road, the center transmits different characteristic parameters to a plurality of vehicles capable of detecting the new feature candidates. The vehicle detects the candidate for the new feature by the detection unit, and matches the feature point indicated by the feature parameter received from the center with respect to the detected candidate for the new feature. Determining whether or not, and transmitting the determination result to the center, the center determines the feature point of the candidate for the new feature based on the determination result received from the plurality of vehicles, the determined The feature is stored in the storage unit by identifying the new feature based on a feature point of a candidate for a new feature, associating the identified new feature with position information of the new feature. It is characterized by updating information.

  In the feature information collection system, the storage unit stores the feature parameters in a hierarchical manner according to the common items of the feature points, and the center is a candidate for a new feature newly established on the road. If there is, for each of the plurality of vehicles that can detect the candidate for the new feature, different feature parameters are transmitted from the common item of the feature parameter hierarchy stored in the storage unit, Whether the new feature candidate is detected by the detection unit, and whether or not the detected new feature candidate matches the feature point of the common item of the hierarchy indicated by the feature parameter received from the center And the determination result is transmitted to the center, and the center determines the feature point of the candidate for the new feature from the common item of the hierarchy based on the determination result received from the plurality of vehicles. The determination process to be determined is repeated for each layer, the new feature is identified based on the feature points of the candidate for the new feature determined corresponding to each layer by the determination process, and the identified new land It is preferable that the feature information stored in the storage unit is updated by associating a feature with the position information of the new feature.

  In the feature information collection system, the center is associated with the determined feature point of the new feature candidate after determining the feature point of the candidate of the new feature from the common item of the hierarchy by the determination process. Preferably, the determination process is executed in the next hierarchy by transmitting different characteristic parameters selected from the common items in the next hierarchy to a plurality of vehicles capable of detecting the candidate for the new feature. .

  In the feature information collection system, the center corresponds to the determined feature point of the candidate for the new feature after determining the feature point of the candidate for the new feature from the common item of the hierarchy by the determination process. The same feature parameter is transmitted to a plurality of vehicles capable of detecting the candidate for the new feature, and the vehicle detects the candidate for the new feature by the detection unit, and the detected new feature It is determined whether or not the candidate matches a feature point indicated by the feature parameter received from the center, and a determination result is transmitted to the center. The center receives the determination result received from the plurality of vehicles. And confirming whether the feature point of the candidate for the new feature determined from the common item of the hierarchy by the determination process is a correct result, and associating it with the feature point of the confirmed candidate for the new feature The respective different feature parameters selected from a common section of the next layer, it is preferable to transmit the candidate of the new feature for a plurality of vehicles detectable.

  In the feature information collection system, it is preferable that each of the layers includes any one of a color of the feature, a shape of the feature, and a template of the feature as a common term.

  In the feature information collection system, the center recognizes a recognition range in which the detection unit recognizes the candidate for the new feature from a recognition region that the detection unit can recognize for a plurality of vehicles traveling on the road. When the detection unit detects the candidate for the new feature within the recognition range specified by the recognition range specification information received from the center, The presence information indicating that there is a candidate for a new feature is transmitted to the center, and when the center receives the presence information from the vehicle, the center determines that there is a candidate for a new feature newly established on the road. It is preferable to do.

  The center of the present invention is communicably connected to a plurality of vehicles having at least a detection unit that detects a feature installed in the traveling direction of the road, and associates the feature with position information of the feature. And a center having at least a storage unit for storing feature parameters corresponding to a plurality of feature points specifying the feature, the center being a candidate for a new feature newly established on the road Are present, the feature parameters of the new feature candidate are transmitted to the plurality of vehicles capable of detecting the candidate for the new feature, respectively, and based on the determination result received from the plurality of vehicles. Determining a point, identifying the new feature based on the determined feature point of the candidate for the new feature, associating the identified new feature with position information of the new feature, and storing the memory In the section The determined feature information is updated, and the determination result is received from the center with respect to the detected new feature candidate when the detection unit detects the new feature candidate in the vehicle. It is determined whether or not the feature point indicated by the feature parameter matches, and the result is transmitted to the center.

  In the feature information collecting method of the present invention, a plurality of vehicles having at least a detection unit that detects a feature installed in a traveling direction of a road are associated with the feature and position information of the feature. A center having at least a storage unit that stores feature information and feature parameters corresponding to a plurality of feature points that specify the feature, and the plurality of vehicles and the center are connected to be communicable with each other. A feature information collection method executed in a feature information collection system, wherein there are a plurality of new feature candidates that can be detected when there are new feature candidates newly established on the road in the center. Different characteristic parameters are transmitted to the vehicle, and the new feature candidate is detected by the detection unit in the vehicle, and the detected new feature candidate is detected with respect to the detected new feature candidate. It is determined whether or not the feature point indicated by the feature parameter received from the vehicle matches, and a determination result is transmitted to the center, and the new location is determined based on the determination result received from the plurality of vehicles at the center. A feature point of an object candidate is determined, the new feature is identified based on the determined feature point of the candidate of the new feature, and the identified new feature and position information of the new feature And the feature information stored in the storage unit is updated.

  The feature information collecting method of the present invention is connected to a plurality of vehicles having at least a detection unit for detecting a feature installed in the traveling direction of a road so as to be communicable, and the location information of the feature and the feature And a feature information collecting method that is executed in a center having at least a storage unit that stores feature information corresponding to a plurality of feature points that identify the feature. When there are new feature candidates newly established on the road, different characteristic parameters are transmitted to the plurality of vehicles capable of detecting the new feature candidates and received from the plurality of vehicles. The feature point of the candidate for the new feature is determined based on the determination result, the new feature is specified based on the feature point of the candidate for the new feature determined, and the specified new feature The feature information stored in the storage unit is updated in association with the position information of the new feature, and the determination result is that the detection unit detects the candidate for the new feature in the vehicle. The detected result is a result transmitted to the center after it is determined whether the detected feature candidate matches the feature point indicated by the feature parameter received from the center. And

  The feature information collection system, the center, and the feature information collection method according to the present invention have an effect of specifying a new feature while reducing the processing load on the vehicle.

FIG. 1 is a diagram illustrating a relationship between a center and a plurality of vehicles according to the present embodiment. FIG. 2 is a diagram illustrating an example of the configuration of the feature information collection system according to the present embodiment. FIG. 3 is a diagram illustrating an example of the characteristic parameter according to the present embodiment. FIG. 4 is a flowchart showing an example of basic processing of the feature information collection system in the present embodiment. FIG. 5 is a flowchart showing another example of the basic processing of the feature information collection system in the present embodiment. FIG. 6 is a diagram illustrating an example of operations until new features executed by the center and a plurality of vehicles according to the present embodiment are specified. FIG. 7 is a diagram illustrating an example of an operation until a new feature executed by the center and the plurality of vehicles according to the present embodiment is specified. FIG. 8 is a diagram illustrating an example of operations until new features executed by the center and a plurality of vehicles according to the present embodiment are specified. FIG. 9 is a diagram illustrating an example of operations until new features executed by the center and a plurality of vehicles according to the present embodiment are specified. FIG. 10 is a diagram illustrating an example of an operation until a new feature executed by the center and the plurality of vehicles according to the present embodiment is specified. FIG. 11 is a diagram illustrating an example of an operation until a new feature executed by the center and the plurality of vehicles according to the present embodiment is specified. FIG. 12 is a diagram illustrating an example of an operation until a new feature executed by the center and the plurality of vehicles according to the present embodiment is specified. FIG. 13 is a flowchart showing details of processing of the feature information collection system corresponding to FIGS. 6 to 12. FIG. 14 is a flowchart showing details of processing of the feature information collection system corresponding to FIGS. 6 to 12. FIG. 15 is a diagram illustrating an example of characteristic parameters allocated to a plurality of vehicles. FIG. 16 is a diagram illustrating another example of characteristic parameters allocated to a plurality of vehicles.

  Hereinafter, embodiments of a feature information collection system and a feature information collection method according to the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment. In addition, constituent elements in the following embodiments include those that can be easily assumed by those skilled in the art or those that are substantially the same.

Embodiment
A configuration of a feature information collection system according to an embodiment of the present invention will be described with reference to FIGS. 1 to 3. FIG. 1 is a diagram illustrating a relationship between a center and a plurality of vehicles according to the present embodiment. FIG. 2 is a diagram illustrating an example of the configuration of the feature information collection system according to the present embodiment. FIG. 3 is a diagram illustrating an example of the characteristic parameter according to the present embodiment.

  As shown in FIG. 1, the feature information collection system according to the present embodiment is roughly mounted on each of a plurality of vehicles A to C (hereinafter sometimes simply referred to as “vehicles”) traveling on a road. The plurality of vehicle-mounted devices 100-1 to 100-3 and the center 200 that is communicably connected to the plurality of vehicle-mounted devices 100-1 to 100-3. In the situation shown in FIG. 1, among the plurality of vehicles A to C, the vehicle A is traveling ahead of the vehicles B and C, and the vehicle B is traveling ahead of the vehicle C. Here, each of the plurality of vehicle-mounted devices 100-1 to 100-3 includes at least detection units 110-1 to 110-3 that detect a feature installed at a destination on the road. Further, the center 200 includes at least a storage unit 206 that stores feature information that associates the feature with position information of the feature, and feature parameters that correspond to a plurality of feature points that specify the feature. Details of each part of the in-vehicle device 100 and the center 200 will be described later with reference to FIG.

  As shown in FIG. 2, the feature information collection system of the present embodiment is configured such that a center 200 and an in-vehicle device 100 are communicably connected. Here, the communication includes remote communication such as wireless communication via the network 300. In addition, each part of these feature information collection systems is communicably connected via an arbitrary communication path.

  In the feature information collection system of the present embodiment, the center 200 as the feature information collection device schematically includes at least a control unit 202 and a storage unit 206. In addition, the in-vehicle device 100 as the new feature monitoring apparatus includes at least a position acquisition unit 112, an output unit (display unit 114 and voice output unit 118), an input unit 116, a control unit 102, and a detection unit 110.

[Configuration of Center 200]
Here, in FIG. 2, when there is a new feature candidate newly established on the road, the center 200 applies to the in-vehicle devices 100 respectively mounted on a plurality of vehicles capable of detecting the new feature candidate. Have a function of transmitting different characteristic parameters. Further, the center 200 determines feature points of new feature candidates based on the determination results received from the in-vehicle devices 100 mounted on a plurality of vehicles, and based on the determined feature points of the new feature candidates. A new feature is specified, the specified new feature is associated with the position information of the new feature, and the feature information stored in the storage unit 206 is updated.

  The center 200 is connected to the vehicle-mounted device 100 via the network 300 via the communication control interface unit 204 so as to be communicable with each other, and includes a control unit 202 and a storage unit 206. The control unit 202 is a control unit that performs various processes. The communication control interface unit 204 is an interface connected to a wireless communication device (not shown) such as an antenna, and has a function of performing communication control between the center 200 and the network 300. That is, the communication control interface unit 204 has a function of communicating data with the in-vehicle device 100 through remote communication such as wireless communication.

  In FIG. 2, a storage unit 206 is a fixed disk device such as an HDD (Hard Disk Drive) and storage means such as an SSD (Solid State Drive), and various databases and tables (feature information database 206a and feature parameter database 206b). Etc.).

  Among these constituent elements of the storage unit 206, the feature information database 206a is a feature information storage unit that stores feature information in which a feature is associated with position information of the feature. Specifically, the feature information database 206a stores feature information indicating features of the feature in association with position information indicating the installation point of the feature. The position information is information indicating the latitude, longitude, and altitude of the feature. The feature information is information indicating feature points that can identify the target feature. By way of example, features include, but are not limited to signs, signs, bus lights, reflectors, road paint, traffic lights, tunnels, piers, bridge girders, pedestrian bridges, and the like. The feature information database 206a stores these pieces of information in advance, and updates the stored feature information based on information on new feature candidates transmitted from a plurality of vehicles.

  The feature parameter database 206b is feature parameter storage means for storing feature parameters corresponding to a plurality of feature points that specify features. Specifically, the feature parameter database 206b stores the feature parameters arranged hierarchically according to the common items of the feature points. Each layer includes any one of a feature color, a feature shape, and a feature template as a common item. Here, an example of the characteristic parameter according to the present embodiment will be described with reference to FIG.

  As shown in FIG. 3, as an example, the feature parameter database 206b organizes and stores the first feature parameters indicating the color of the feature in the first layer that is the highest layer. The first feature parameter is, for example, a parameter arranged in the first hierarchy using the feature color as a common term among a plurality of feature points specifying the feature. Examples of feature colors include, but are not limited to, red, green, blue, and the like. Further, the feature parameter database 206b organizes and stores the second feature parameters indicating the shape of the feature in the second layer that is the next layer after the first layer. The second feature parameter is, for example, a parameter arranged in the second hierarchy with the shape of the feature among the plurality of feature points specifying the feature as a common term. Examples of the shape of the feature include, but are not limited to, a rectangle, a diamond shape, a circle, an ellipse, and other shapes. Also, the feature parameter database 206b organizes and stores the third feature parameters indicating the feature templates in the third layer, which is the next layer after the second layer. The third feature parameter is, for example, a parameter arranged in the third hierarchy using a feature template as a common item among a plurality of feature points specified as a feature. Examples of feature templates include, but are not limited to, signage templates such as M hamburgers, I-caching, E-gas stations, and T car showrooms. In the example of FIG. 3, the first feature parameter may be a feature parameter indicating the shape of the feature, and the second feature parameter may be a feature parameter indicating the color of the feature.

  In the present embodiment, the feature parameters stored in the feature parameter database 206b are transmitted from the center 200 to the in-vehicle device 100 mounted on each of a plurality of vehicles capable of detecting new feature candidates. In response to this, the in-vehicle device 100 transmits a determination result indicating whether the new feature candidate detected by the detection unit 110 matches the feature point indicated by the feature parameter received from the center 200 to the center 200. To do. Thereby, the center 200 specifies a new feature based on the determination result received from the vehicle-mounted devices 100 of a plurality of vehicles, and newly registers the feature information related to the specified new feature in the feature information database 206a. , Feature information can be updated.

  Returning to FIG. 2, the description of the configuration of the feature information collection system will be continued. The recognition range designation information file 206c is a recognition range designation that designates a recognition range in which the detection unit 110 recognizes a candidate for a new feature from recognition areas that can be recognized by the detection unit 110 provided in each of the plurality of vehicles traveling on the road. It is a recognition range designation information storage means for storing information. The recognition region of the detection unit 110 is a region that can be recognized to the maximum by the detection unit 110 determined according to the angle of view, performance, and the like of the detection unit 110. The recognition range is a partial range within this recognition area. For example, when the recognition area is a horizontally long rectangular area, the recognition range includes a predetermined range such as upper left, upper center, upper right, lower left, lower center, and lower right of the recognition area, but is not limited thereto. In the present embodiment, the recognition range designation information stored in the recognition range designation information file 206c is used when the center 200 causes the detection unit 110 included in each of the plurality of vehicles traveling on the road to recognize new feature candidates. It is used to limit the recognition range so as to reduce the processing load of the detection unit 110 of the vehicle.

  In addition, the storage unit 206 may include a road network data storage unit that stores road network data that defines a road network, and a map data storage unit that stores map data. In the present embodiment, the road network data stored in the road network data storage unit and the map data stored in the map data storage unit together with the feature information stored in the feature information database 206a, Used for driving assistance.

  The control unit 202 has an internal memory for storing a control program such as an OS (Operating System), a program that defines various processing procedures, and necessary data. And the control part 202 performs the information processing for performing various processes with these programs. The control unit 202 includes a transmission unit 202a, a reception unit 202b, and an update unit 202c in terms of functional concept.

  Among these, the transmission part 202a is a transmission means which transmits information via the communication control interface part 204 with respect to the vehicle equipment 100 respectively mounted in the some vehicle connected so that communication with the center 200 was possible. . In the present embodiment, when there is a new feature candidate newly installed on the road, the transmission unit 202a performs the on-vehicle device 100 mounted on each of the plurality of vehicles capable of detecting the new feature candidate. Different characteristic parameters stored in the characteristic parameter database 206b are transmitted. In addition, the transmission unit 202a recognizes a candidate for a new feature from a recognition area that can be recognized by the detection unit 110 with respect to the in-vehicle device 100 mounted on each of a plurality of vehicles traveling on the road. The recognition range designation information stored in the recognition range designation information file 206c for designating the recognition range to be transmitted is transmitted.

  The receiving unit 202b is a receiving unit that receives information from the in-vehicle device 100 mounted on each of a plurality of vehicles connected to the center 200 via the communication control interface unit 204. In the present embodiment, the receiving unit 202b receives a determination result for determining whether or not the feature point indicated by the feature parameter matches from the in-vehicle device 100 mounted on each of the plurality of vehicles to which the feature parameter has been transmitted. In addition, the receiving unit 202b detects new feature candidates by the detection unit 110 within the recognition range designated by the recognition range designation information from the in-vehicle devices 100 mounted on the plurality of vehicles to which the recognition range designation information is transmitted. Presence information indicating that there is a candidate for the new feature to be transmitted is received. Here, the presence information may include vehicle position information acquired by a vehicle position acquisition unit 102d described later when the in-vehicle device 100 detects a new feature by the detection unit 110, or may be described below. It may include position information of new feature candidates calculated by the position calculation unit 102e.

  The update unit 202c is an update unit that updates the feature information stored in the feature information database 206a. In the present embodiment, the updating unit 202c determines feature points of new feature candidates based on the determination result received by the receiving unit 202b from the in-vehicle device 100 mounted on each of a plurality of vehicles. Then, the update unit 202c identifies the new feature based on the determined feature points of the new feature candidate. Then, the update unit 202c updates the feature information stored in the feature information database 206a by associating the identified new feature with the position information of the new feature.

[Configuration of in-vehicle device 100]
In FIG. 2, the in-vehicle device 100 detects a new feature candidate using the detection unit 110, and whether the detected new feature candidate matches the feature point indicated by the feature parameter received from the center 200. A function of determining whether or not to transmit the determination result to the center 200.

  As shown in FIG. 2, the in-vehicle device 100 includes a position acquisition unit 112 having a GPS function so that the current position can be acquired in real time. The in-vehicle device 100 includes an input unit 116 and an output unit including at least a display unit 114 and an audio output unit 118. The in-vehicle device 100 includes a detection unit 110 such as a camera capable of capturing a still image and a moving image of a destination of a vehicle and a radar capable of measuring a distance from a target object of the destination of the vehicle.

  Here, the display unit 114 is configured by display means (for example, liquid crystal or organic EL) that displays a display screen of an application or the like, feature information regarding existing features or new features, current position information, route guidance information, and the like. Display and monitor). The sound output unit 118 is sound output means (for example, a speaker) that outputs sound data as sound. The in-vehicle device 100 includes an input unit 116 (for example, a key input unit, a touch panel, and a microphone) that inputs and operates various conditions. The input / output control interface unit 108 controls the position acquisition unit 112, the display unit 114, the input unit 116, the audio output unit 118, the detection unit 110, and the like.

  In the present embodiment, the position acquisition unit 112 may be a position acquisition unit that receives a position information signal transmitted from the position transmission device 500, for example. Here, the position transmitting device 500 may be a GPS device that transmits a position information signal (GPS signal). Further, the position acquisition unit 112 may receive other sanitary signals (for example, Galileo satellite signals).

  The communication control interface unit 104 is an interface connected to a wireless communication device (not shown) such as an antenna, and has a function of performing communication control between the in-vehicle device 100 and the network 300. That is, the communication control interface unit 104 has a function of communicating data with the center 200 or the like via remote communication such as wireless communication.

  The control unit 102 includes an internal memory for storing a control program such as an OS, a program defining various processing procedures, and necessary data. And the control part 102 performs the information processing for performing various processes by these programs. The control unit 102 includes a reception unit 102a, an image analysis unit 102b, a transmission unit 102c, a vehicle position acquisition unit 102d, and a relative position calculation unit 102e in terms of functional concept.

  Among these, the receiving unit 102 a is a receiving unit that receives information from the center 200 that is communicably connected to the in-vehicle device 100 via the communication control interface unit 104. In the present embodiment, the receiving unit 102a receives a characteristic parameter transmitted from the center 200. The receiving unit 102a receives the recognition range designation information transmitted from the center 200.

  The image analysis unit 102b is an image analysis unit that analyzes an image captured by the detection unit 110. In the present embodiment, the image analysis unit 102b detects a new feature candidate using the detection unit 110, and whether the detected new feature candidate matches the feature point indicated by the feature parameter received from the center 200. Determine whether or not. Further, the image analysis unit 102b detects a new feature candidate by performing image analysis on the image within the recognition range specified by the recognition range specification information received from the center 200 by the reception unit 102a.

  The transmission unit 102 c is a transmission unit that transmits information to the center 200 that is communicably connected to the in-vehicle device 100 via the communication control interface unit 104. In the present embodiment, the transmission unit 102c transmits to the center 200 the determination result regarding the new feature candidate determined by the image analysis unit 102b. In addition, when the detection unit 110 detects a new feature candidate within the recognition range specified by the recognition range specification information by the processing of the image analysis unit 102b, the transmission unit 102c indicates that the new feature candidate exists. Is transmitted to the center 200.

  The vehicle position acquisition unit 102d is vehicle position acquisition means for acquiring current position information of a vehicle on which the in-vehicle device 100 is mounted. In the present embodiment, the vehicle position acquisition unit 102d may acquire the position information calculated from the position information signal received from the position transmission device 500 by the position acquisition unit 112 as the current position information of the vehicle. Here, the vehicle position acquisition unit 102d may acquire the current position information of the vehicle every predetermined time (predetermined period).

  The relative position calculation unit 102e is a relative position calculation unit that calculates a relative position of a new feature or a new feature candidate installed in the traveling direction of the vehicle with respect to the current position of the vehicle. In the present embodiment, the relative position calculation unit 102e is a distance from the current position of the vehicle based on the current position information of the vehicle acquired by the vehicle position acquisition unit 102d to the new feature measured by the detection unit 110 such as a radar. Based on the above, the relative position of the new feature with respect to the current position of the vehicle is calculated to obtain the position information of the new feature.

  In addition, the in-vehicle device 100 may further include a storage unit. The storage means is a storage means such as a large-capacity storage means (for example, HDD and SSD) and / or a small-capacity high-speed memory (for example, cache memory) configured using SRAM (Static Random Access Memory) or the like. Various databases, files, and tables may be stored. Here, the storage means may temporarily store various files and the like.

[Processing of feature information collection system]
Next, an example of the processing of the feature information collection system according to the present embodiment configured as described above will be described in detail with reference to FIGS. Here, FIG. 4 is a flowchart showing an example of basic processing of the feature information collection system in the present embodiment. FIG. 5 is a flowchart showing another example of the basic processing of the feature information collection system in the present embodiment. 6-12 is a figure which shows an example of the operation | movement until it specifies the new feature performed with the center concerning this embodiment, and a some vehicle. FIGS. 13 and 14 are flowcharts showing details of processing of the feature information collection system corresponding to FIGS. 6 to 12.

  As shown in FIG. 4, first, the transmission unit 202a of the center 200 detects the vehicle-mounted device 100 mounted on each of a plurality of vehicles traveling on the road from the recognition area that can be recognized by the detection unit 110. The recognition range designation information stored in the recognition range designation information file 206c for designating a recognition range in which the unit 110 recognizes a new feature candidate is transmitted (step SA-1). For example, in step SA-1, the transmission unit 202a is configured to detect each of the detection units 110-1 to 110-3 with respect to the in-vehicle devices 100-1 to 3 mounted on the plurality of vehicles A to C illustrated in FIG. When the recognition area is a horizontally long rectangular area, recognition range designation information for designating a predetermined range at the upper left of the recognition area is transmitted.

  The receiving unit 102a of the in-vehicle device 100 receives the recognition range designation information transmitted from the center 200 by the processing of the transmitting unit 202a in step SA-1 (step SA-2).

  The image analysis unit 102b of the in-vehicle device 100 performs image analysis on the image within the recognition range designated by the recognition range designation information received from the center 200 by the processing of the reception unit 102a in step SA-2, thereby performing a new location. Detect object candidates. Then, the image analysis unit 102b determines whether a new feature candidate is detected based on the image analysis result (step SA-3). For example, in step SA-3, the image analysis unit 102b pays attention only to an image within a recognition range corresponding to a predetermined range on the upper left in the recognition region of the detection unit 110, and performs image analysis on a change in the image within this recognition range. By detecting by this, a candidate for a new feature is detected.

  If it is determined in step SA-3 that a candidate for a new feature has been detected by the process of the image analysis unit 102b (step SA-3: Yes), the process proceeds to step SA-4. On the other hand, when it is determined in step SA-3 that the candidate for the new feature has not been detected by the processing of the image analysis unit 102b (step SA-3: No), the image analysis unit 102b performs the processing of step SA-3. Repeat until it is determined that a new feature candidate has been detected.

  When the transmission unit 102c of the vehicle-mounted device 100 detects a new feature candidate by the detection unit 110 within the recognition range designated by the recognition range designation information by the processing of the image analysis unit 102b in step SA-3, the new land Presence information indicating that there is an object candidate is transmitted to the center 200 (step SA-4).

  The receiving unit 202b of the center 200 selects a new feature candidate by the detection unit 110 within the recognition range specified by the recognition range specification information from the in-vehicle devices 100 mounted on the plurality of vehicles to which the recognition range specification information is transmitted. Presence information indicating that there is a candidate for the new feature transmitted when it is detected is received (step SA-5). For example, in step SA-5, the reception unit 202b transmits the presence information transmitted by the processing of the transmission unit 102c of the in-vehicle devices 100-1 to 3 mounted on the plurality of vehicles A to C traveling on the same road. Receive.

  When the transmission unit 202a of the center 200 receives the presence information by the processing of the reception unit 202b in step SA-5, the transmission unit 202a determines that there is a new feature candidate newly established on the road (step SA-6).

  As determined in step SA-6, the transmission unit 202a of the center 200 is installed in each of a plurality of vehicles capable of detecting the new feature candidates when there are new feature candidates newly installed on the road. The different feature parameters stored in the feature parameter database 206b are transmitted to the in-vehicle device 100 that has been made (step SA-7). For example, in step SA-7, the transmission unit 202a, for a vehicle-mounted device 100-1 mounted on the vehicle A among a plurality of vehicles A to C traveling on the same road, performs M hamburger and I caching. The characteristic parameter indicating the signboard template is transmitted. Moreover, the transmission part 202a transmits the characteristic parameter which shows the template of the signboard of I caching and E gas station with respect to the vehicle equipment 100-2 mounted in the vehicle B. FIG. Moreover, the transmission part 202a transmits the characteristic parameter which shows the template of the signboard of E gas station and M hamburger with respect to the vehicle equipment 100-3 mounted on the vehicle C.

  In the present embodiment, a plurality of vehicles capable of detecting new feature candidates are determined on the center 200 side based on the vehicle position information and the new feature candidate position information transmitted from the in-vehicle device 100 side. The For example, the control unit 202 of the center 200 has a current position of a vehicle traveling on a road on which a candidate for a new feature as a target is installed on the near side of the position of the candidate for a new feature, and If the distance between the current position of the vehicle and the position of the candidate for the new feature is within the distance that the detection unit 110 can detect the target, the vehicle determines that the candidate for the new feature can be detected.

  The receiving unit 102a of the in-vehicle device 100 receives the characteristic parameter transmitted from the center 200 by the processing of the transmitting unit 202a in step SA-7 (step SA-8). For example, in step SA-8, the receiving unit 102a of the in-vehicle device 100-1 mounted on the vehicle A receives the characteristic parameter indicating the template for the M hamburger and I-caching signboard. In addition, the receiving unit 102a of the in-vehicle device 100-2 mounted on the vehicle B receives the characteristic parameters indicating the I caching and E gas station sign templates. Moreover, the receiving part 102a of the vehicle equipment 100-3 mounted in the vehicle C receives the characteristic parameter which shows the template of the signboard of E gas station and M hamburger.

  The image analysis unit 102b of the in-vehicle device 100 indicates the feature parameter received from the center 200 by the processing of the reception unit 102a in step SA-8 with respect to the new feature candidate detected by the detection unit 110 in step SA-3. It is determined whether or not the feature point matches (step SA-9). For example, in step SA-9, the image analysis unit 102b of the in-vehicle device 100-1 mounted on the vehicle A determines whether the new feature candidate matches the template of the M hamburger or I-caching signboard. Judge by matching. In addition, the image analysis unit 102b of the vehicle-mounted device 100-2 mounted on the vehicle B determines whether a new feature candidate matches the template of the I-caching or E-gas station signboard by template matching. . In addition, the image analysis unit 102b of the in-vehicle device 100-3 mounted on the vehicle C determines whether or not a new feature candidate matches the template of the E gas station or M hamburger sign by template matching. .

  The transmission unit 102c of the in-vehicle device 100 transmits the determination result regarding the new feature candidate determined by the processing of the image analysis unit 102b in Step SA-9 to the center 200 (Step SA-10). For example, in step SA-10, the transmission unit 102c of the vehicle-mounted device 100-1 mounted on the vehicle A matches the template for the M hamburger signboard with the candidate for the new feature, Transmits a determination result indicating that they do not match to the center 200. In addition, the transmission unit 102c of the vehicle-mounted device 100-2 mounted on the vehicle B determines that the new feature candidate does not match either the I-caching signboard template or the E-gas station signboard template. Is transmitted to the center 200. In addition, the transmission unit 102c of the in-vehicle device 100-3 mounted on the vehicle C has a new feature candidate that does not match the template of the E gas station signboard, but matches the template of the M hamburger signboard. A determination result indicating that is transmitted to the center 200. Thereafter, the in-vehicle device 100 ends the process.

  The receiving unit 202b of the center 200 receives a determination result for determining whether or not the feature point matches the feature point indicated by the feature parameter from each of the in-vehicle devices 100 mounted on the plurality of vehicles to which the feature parameter is transmitted (step SA). -11). For example, in step SA-11, the reception unit 202b receives the same new feature transmitted in step SA-10 by the processing of the transmission unit 102c of the in-vehicle devices 100-1 to 3 mounted on the vehicles A to C, respectively. A plurality of determination results regarding the candidate are received.

  The update unit 202c of the center 200 is a feature point of a candidate for a new feature based on a plurality of determination results received by the processing of the receiving unit 202b in step SA-11 from the in-vehicle device 100 mounted on each of a plurality of vehicles. Is determined (step SA-12). For example, in Step SA-12, the update unit 202c transmits the vehicle A transmitted in Step SA-10 by the processing of the transmission unit 102c of the in-vehicle devices 100-1 to 3 mounted on the plurality of vehicles A to C, respectively. Based on the determination result, the determination result of the vehicle B, and the determination result of the vehicle C, the feature point of the candidate for the new feature is determined. Specifically, the determination result of the vehicle A and the determination result of the vehicle C indicate that the new feature candidate matches the template of the M hamburger signboard. As candidate feature points.

  The update unit 202c of the center 200 identifies the new feature based on the feature points of the new feature candidate determined in Step SA-12 (Step SA-13). For example, in step SA-13, the update unit 202c identifies the new feature as an M hamburger sign based on the template for the M hamburger sign determined as the feature point of the new feature candidate.

  The update unit 202c of the center 200 updates the feature information stored in the feature information database 206a by associating the new feature specified in Step SA-13 with the position information of the new feature (Step SA). -14). For example, in step SA-14, the updating unit 202c updates the feature information stored in the feature information database 206a by associating the M hamburger with the position information of the M hamburger. Thereafter, the center 200 ends the process.

  In the present embodiment, the position information of the new feature is acquired by processing of the vehicle position acquisition unit 102d and the relative position calculation unit 102e of the in-vehicle device 100. For example, when a new feature candidate is detected, the vehicle position acquisition unit 102d of the in-vehicle device 100 acquires the current position information of the vehicle. The relative position calculation unit 102e of the in-vehicle device 100 then determines the current position of the vehicle based on the distance from the current position of the vehicle based on the current position information of the vehicle to the new feature measured by the detection unit 110 such as a radar. The relative position of the new feature with respect to is calculated, and the position information of the new feature is acquired. Then, the transmission unit 102c of the in-vehicle device 100 transmits the position information of the new feature to the center 200. Note that the timing for acquiring position information of a new feature is not limited to the point in time when a new feature candidate is detected. In the present embodiment, when the center 200 identifies a new feature, the center 200 acquires position information of the new feature and transmits it to the center 200 to the in-vehicle device 100 mounted on the vehicle capable of detecting the new feature. By instructing, the position information of the new feature may be acquired.

  Thus, in the feature information collection system according to the present embodiment, when there are new feature candidates newly established on the road, the center 200 detects a plurality of vehicles that can detect the new feature candidates. Different characteristic parameters are transmitted. Then, the vehicle detects a candidate for a new feature by the detection unit 110, determines whether or not the detected feature candidate matches the feature point indicated by the feature parameter received from the center 200, The determination result is transmitted to the center 200. Then, the center 200 determines the feature points of the new feature candidate based on the determination results received from the plurality of vehicles, specifies the new feature based on the determined feature point of the new feature candidate, The feature information stored in the feature information database 206a is updated by associating the new feature with the position information of the new feature. Accordingly, the feature information collection system according to the present embodiment can perform processing for specifying new features by distributed processing by the center 200 allocating different feature parameters to a plurality of vehicles. A new feature can be specified while reducing the processing load on the vehicle.

  Here, in the process shown in FIG. 4, an example in which templates of different features are used as feature parameters has been described. However, since template matching is a process with a relatively high processing load, the processing load per unit is further increased. In order to reduce it, as shown in FIG. 5, it is desirable to perform a process that does not execute a vehicle other than the one that requires template matching. Hereinafter, the process in this case will be described with reference to FIG.

  In FIG. 5, the processing of Step SB-1 to Step SB-6 is the same as the processing of Step SA-1 to Step SA-6 of FIG.

  As shown in FIG. 5, the transmission unit 202a of the center 200 can detect a new feature candidate when there is a new feature candidate newly established on the road, as determined in step SB-6. Of the feature parameters stored in the feature parameter database 206b, different feature parameters are transmitted from the common item in the upper hierarchy to the vehicle-mounted devices 100 respectively mounted on a plurality of vehicles (step SB-7). For example, in step SB-7, the transmission unit 202a transmits the features of the on-vehicle devices 100-1 to 100-3 mounted on the plurality of vehicles A to C traveling on the same road illustrated in FIG. Different first characteristic parameters arranged in the first hierarchy are transmitted using the color as a common term. Specifically, the transmission unit 202a transmits the first characteristic parameter indicating red to the in-vehicle device 100-1 mounted on the vehicle A. Moreover, the transmission part 202a transmits the 1st characteristic parameter which shows green with respect to the vehicle equipment 100-2 mounted in the vehicle B. FIG. Moreover, the transmission part 202a transmits the 1st characteristic parameter which shows blue with respect to the vehicle equipment 100-3 mounted in the vehicle C. FIG.

  The receiving unit 102a of the in-vehicle device 100 receives the characteristic parameter transmitted from the center 200 by the processing of the transmitting unit 202a in step SB-7 (step SB-8). For example, in step SB-8, the receiving unit 102a of the in-vehicle device 100-1 mounted on the vehicle A receives the first characteristic parameter indicating red. The receiving unit 102a of the in-vehicle device 100-2 mounted on the vehicle B receives the first characteristic parameter indicating green. Moreover, the receiving part 102a of the vehicle equipment 100-3 mounted in the vehicle C receives the 1st characteristic parameter which shows blue.

  The image analysis unit 102b of the in-vehicle device 100 detects a candidate for a new feature by the detection unit 110 (step SB-9). For example, in step SB-9, the detection units 110-1 to 110-3 mounted on the vehicles A to C respectively detect the same new feature candidates installed in the traveling direction.

  The image analysis unit 102b of the in-vehicle device 100 indicates the feature parameter received from the center 200 by the processing of the reception unit 102a in step SB-8 with respect to the new feature candidate detected by the detection unit 110 in step SB-9. It is determined whether or not the feature point of the common item in the upper hierarchy matches (step SB-10). For example, in step SB-10, the image analysis unit 102b of the vehicle-mounted device 100-1 mounted on the vehicle A performs image analysis to determine whether a new feature candidate matches the red color indicated by the first feature parameter. Judgment by In addition, the image analysis unit 102b of the in-vehicle device 100-2 mounted on the vehicle B determines whether or not a new feature candidate matches the green color indicated by the first feature parameter by image analysis. Further, the image analysis unit 102b of the in-vehicle device 100-3 mounted on the vehicle C determines whether or not a new feature candidate matches the blue color indicated by the first feature parameter by image analysis.

  The transmission unit 102c of the in-vehicle device 100 transmits the determination result regarding the new feature candidate determined by the processing of the image analysis unit 102b in Step SB-10 to the center 200 (Step SB-11). For example, in step SB-11, the transmission unit 102c of the vehicle-mounted device 100-1 mounted on the vehicle A displays a determination result indicating that the new feature candidate matches the red color indicated by the first feature parameter. 200. In addition, the transmission unit 102c of the in-vehicle device 100-2 mounted on the vehicle B transmits to the center 200 a determination result indicating that the new feature candidate does not match the green color indicated by the first feature parameter. . In addition, the transmission unit 102c of the in-vehicle device 100-3 mounted on the vehicle C transmits to the center 200 a determination result indicating that the new feature candidate does not match the blue color indicated by the first feature parameter. .

  Whether the reception unit 202b of the center 200 matches the feature point indicated by the upper layer feature parameter from the in-vehicle device 100 mounted on each of the plurality of vehicles to which the upper layer feature parameter is transmitted in step SB-7. The determination result of determining is received (step SB-12). For example, in step SB-12, the receiving unit 202b receives the same new feature transmitted in step SB-11 by the processing of the transmitting unit 102c of the in-vehicle devices 100-1 to 3 mounted on the vehicles A to C, respectively. A plurality of determination results regarding the candidate are received.

  The update unit 202c of the center 200 is newly started from the common item in the upper hierarchy based on the plurality of determination results received from the in-vehicle device 100 mounted on each of the plurality of vehicles by the processing of the reception unit 202b in Step SB-12. Feature points of feature candidates are determined (step SB-13). For example, in step SB-13, the update unit 202c transmits the vehicle A transmitted in step SB-11 by the processing of the transmission unit 102c of the in-vehicle devices 100-1 to 3 mounted on the plurality of vehicles A to C, respectively. Based on the determination result, the determination result of the vehicle B, and the determination result of the vehicle C, the feature point of the candidate for the new feature is determined from the common item in the upper hierarchy. Specifically, the determination result of the vehicle A matches the red of the first feature parameter with the new feature candidate, and the determination result of the vehicle B and the determination result of the vehicle C are the new feature candidate, Since it indicates that the first feature parameter does not match the green and blue colors indicated by the first feature parameter, the feature point relating to the candidate color of the new feature is determined as red.

  In step SB-13, the transmission unit 202a of the center 200 ranks the feature points of new feature candidates based on the determination results received from the in-vehicle devices 100 mounted on a plurality of vehicles by the processing of the update unit 202c. When determined from the common item of the hierarchy, the next hierarchy among the feature parameters stored in the feature parameter database 206b for the in-vehicle device 100 mounted on each of the plurality of vehicles capable of detecting the candidate for the new feature. Different characteristic parameters are transmitted from the common term (step SB-14). For example, in step SB-14, the transmission unit 202a is not shown in FIG. 1 with respect to the in-vehicle devices 100-2 to 3 mounted on the plurality of vehicles B to C traveling on the same road. Transmits different second characteristic parameters arranged in the second hierarchy with the shape of the feature as a common term to the vehicle-mounted device 100-4 mounted on the vehicle D following the vehicle C. Specifically, the transmission unit 202a transmits a second feature parameter indicating a rectangle to the in-vehicle device 100-2 mounted on the vehicle B. Moreover, the transmission part 202a transmits the 2nd characteristic parameter which shows a diamond type | mold with respect to the vehicle equipment 100-3 mounted in the vehicle C. FIG. Moreover, the transmission part 202a transmits the 2nd characteristic parameter which shows a circle | round | yen and an ellipse with respect to the vehicle equipment 100-4 mounted in the vehicle D. FIG.

  In step SB-14, it is assumed that the traveling vehicle A has passed a point where a new feature candidate is installed, and has changed to a state where the new feature candidate cannot be detected. Therefore, the transmission unit 202a does not transmit the second feature parameters arranged in the second hierarchy to the in-vehicle device 100-1 mounted on the vehicle A.

  The receiving unit 102a of the in-vehicle device 100 receives the characteristic parameter transmitted from the center 200 by the processing of the transmitting unit 202a in step SB-14 (step SB-15). For example, in step SB-15, the receiving unit 102a of the in-vehicle device 100-2 mounted on the vehicle B receives the second feature parameter indicating a rectangle. Moreover, the receiving part 102a of the vehicle equipment 100-3 mounted in the vehicle C receives the 2nd characteristic parameter which shows a diamond type. The receiving unit 102a of the in-vehicle device 100-3 mounted on the vehicle D receives the second characteristic parameter indicating a circle and an ellipse.

  The image analysis unit 102b of the in-vehicle device 100 detects a new feature candidate by the detection unit 110 (step SB-16). For example, in Step SB-16, the detection units 110-2 to 110-4 mounted on the vehicles B to D respectively detect the same red new feature candidates installed in the traveling direction.

  The image analysis unit 102b of the in-vehicle device 100 indicates the feature parameter received from the center 200 by the processing of the receiving unit 102a in step SB-15 for the new feature candidate detected by the detecting unit 110 in step SB-16. It is determined whether or not it matches the feature point of the common item in the next hierarchy (step SB-17). For example, in step SB-17, the image analysis unit 102b of the vehicle-mounted device 100-2 mounted on the vehicle B performs image analysis to determine whether the new feature candidate matches the rectangle indicated by the second feature parameter. Judgment by Further, the image analysis unit 102b of the in-vehicle device 100-3 mounted on the vehicle C determines whether or not the new feature candidate matches the diamond type indicated by the second feature parameter. In addition, the image analysis unit 102b of the in-vehicle device 100-4 mounted on the vehicle D determines whether the new feature candidate matches the circle or the ellipse indicated by the second feature parameter by image analysis.

  The transmission unit 102c of the in-vehicle device 100 transmits the determination result regarding the new feature candidate determined by the processing of the image analysis unit 102b in Step SB-17 to the center 200 (Step SB-18). For example, in step SB-18, the transmission unit 102c of the vehicle-mounted device 100-2 mounted on the vehicle B displays a determination result indicating that the new feature candidate matches the rectangle indicated by the second feature parameter. 200. In addition, the transmission unit 102c of the vehicle-mounted device 100-3 mounted on the vehicle C sends a determination result indicating that the new feature candidate does not match the diamond type indicated by the second feature parameter to the center 200. Send. In addition, the transmission unit 102c of the in-vehicle device 100-4 mounted on the vehicle D transmits a determination result indicating that the new feature candidate does not match the circle and the ellipse indicated by the second feature parameter to the center 200. Send.

  Whether the reception unit 202b of the center 200 matches the feature point indicated by the feature parameter of the next layer from the in-vehicle device 100 mounted on each of the plurality of vehicles to which the feature parameter of the next layer is transmitted in Step SB-14. The determination result for determining whether or not is received (step SB-19). For example, in step SB-19, the receiving unit 202b receives the same new red land transmitted in step SB-18 by the processing of the transmitting unit 102c of the vehicle-mounted devices 100-2 to 4-4 mounted on the vehicles B to D, respectively. A plurality of determination results related to the object candidates are received.

  The update unit 202c of the center 200 starts from the common item of the next layer based on the plurality of determination results received from the in-vehicle device 100 mounted on each of the plurality of vehicles by the processing of the reception unit 202b in Step SB-19. A feature point of a candidate for a new feature is determined (step SB-20). For example, in step SB-20, the update unit 202c transmits the vehicle B transmitted in step SB-18 by the processing of the transmission unit 102c of the in-vehicle devices 100-2 to 4 mounted on the plurality of vehicles B to D, respectively. Based on the determination result, the determination result of the vehicle C, and the determination result of the vehicle D, the feature point of the candidate for the new feature is determined from the common item of the next hierarchy. Specifically, the determination result of the vehicle B matches the rectangle indicated by the second feature parameter of the new feature candidate, and the determination result of the vehicle C and the determination result of the vehicle D include the new feature candidate, Since it indicates that the second feature parameter does not match the diamond shape, circle, or ellipse, the feature point related to the shape of the candidate for the new feature is determined as a rectangle.

  The update unit 202c of the center 200 determines whether or not the new feature can be specified based on the feature points of the new feature candidates determined in Step SB-13 and Step SB-20 (Step SB-21). . For example, in step SB-21, can the update unit 202c identify the new feature by referring to the feature parameter base 206b based on the feature points related to the color and shape of the candidate for the new feature determined so far? Determine whether or not. Specifically, the update unit 202c determines that identification is possible when there is a single red and rectangular template indicated by the third feature parameter arranged in the next hierarchy, and there are a plurality of red and rectangular templates. Is determined to be unidentifiable.

  If the update unit 202c of the center 200 determines in step SB-21 that a new feature can be specified (step SB-21: Yes), the update unit 202c proceeds to the next step SB-22. On the other hand, if the update unit 202c of the center 200 determines that the new feature cannot be specified in step SB-21 (step SB-21: No), the process returns to the process of step SB-14.

  Here, the red and rectangular templates indicated by the third characteristic parameter include, for example, a plurality of templates such as M hamburger, I caching, E gas station, and T car showroom as shown in FIG. An example will be described in which it is determined in SB-21 that a new feature cannot be specified by the processing of the update unit 202c, the processing returns to step SB-14, and the processing of steps SB-14 to 21 is repeated again.

  If it is determined in step SB-21 that the update unit 202c determines that a new feature cannot be specified (step SB-21: No), the transmission unit 202a of the center 200 can detect a plurality of candidates for the new feature. Among the feature parameters stored in the feature parameter database 206b, different feature parameters are transmitted from the common item of the next hierarchy to the in-vehicle device 100 mounted on each vehicle (step SB-14). For example, in step SB-14, the transmission unit 202a follows the vehicle C, although not shown in FIG. 1 with respect to the vehicle-mounted device 100-3 mounted on the plurality of vehicles C traveling on the same road. For the vehicle-mounted device 100-4 mounted on the vehicle D and the vehicle-mounted device 100-5 mounted on the vehicle E following the vehicle D, the feature template is used as a common item in the third layer. Send different arranged third feature parameters. Specifically, the transmission unit 202a transmits, to the vehicle-mounted device 100-3 mounted on the vehicle C, a third feature parameter indicating a template for an M hamburger and an I-caching signboard. Moreover, the transmission part 202a transmits the 3rd characteristic parameter which shows the template of the signboard of I caching and E gas station with respect to the vehicle equipment 100-4 mounted in the vehicle D. FIG. Moreover, the transmission part 202a transmits the 3rd characteristic parameter which shows the signboard template of E gas station and T car showroom with respect to the vehicle equipment 100-5 mounted in the vehicle E. FIG.

  In step SB-14, it is assumed that the traveling vehicle A and vehicle B have passed through a point where a new feature candidate has been installed, and the new feature candidate cannot be detected. Yes. For this reason, the transmission unit 202a does not transmit the third feature parameters arranged in the third hierarchy to the in-vehicle devices 100-1 and 2 mounted on the vehicles A and B, respectively.

  The receiving unit 102a of the in-vehicle device 100 receives the characteristic parameter transmitted from the center 200 by the processing of the transmitting unit 202a in step SB-14 (step SB-15). For example, in step SB-15, the receiving unit 102a of the in-vehicle device 100-3 mounted on the vehicle C receives the third characteristic parameter indicating the template for the M hamburger and the I-caching signboard. In addition, the receiving unit 102a of the in-vehicle device 100-4 mounted on the vehicle D receives the third characteristic parameter indicating the template of the sign for the I caching and the E gas station. Moreover, the receiving part 102a of the vehicle equipment 100-5 mounted in the vehicle E receives the 3rd characteristic parameter which shows the template of the signboard of E gasoline station and T car showroom.

  The image analysis unit 102b of the in-vehicle device 100 detects a new feature candidate by the detection unit 110 (step SB-16). For example, in step SB-16, the detection units 110-3 to 110-5 mounted on the vehicles C to E respectively detect candidates for the same red rectangular new feature installed in the traveling direction.

  The image analysis unit 102b of the in-vehicle device 100 indicates the feature parameter received from the center 200 by the processing of the receiving unit 102a in step SB-15 for the new feature candidate detected by the detecting unit 110 in step SB-16. It is determined whether or not it matches the feature point of the common item in the next hierarchy (step SB-17). For example, in step SB-17, the image analysis unit 102b of the vehicle-mounted device 100-3 mounted on the vehicle C determines that the new feature candidate is a template for an M hamburger or I-caching signboard indicated by the third feature parameter. Whether or not they match is determined by template matching. In addition, the image analysis unit 102b of the vehicle-mounted device 100-4 mounted on the vehicle D determines whether the new feature candidate matches the I caching or E gas station signboard template indicated by the third feature parameter. Is determined by template matching. In addition, the image analysis unit 102b of the in-vehicle device 100-5 mounted on the vehicle E determines whether the new feature candidate matches the template of the E gas station or T car showroom sign indicated by the third characteristic parameter. Is determined by template matching.

  The transmission unit 102c of the in-vehicle device 100 transmits the determination result regarding the new feature candidate determined by the processing of the image analysis unit 102b in Step SB-17 to the center 200 (Step SB-18). For example, in step SB-18, the transmission unit 102c of the vehicle-mounted device 100-3 mounted on the vehicle C matches the template of the M hamburger sign indicated by the third feature parameter. A determination result indicating that the I-caching signboard template did not match is transmitted to the center 200. In addition, the transmission unit 102c of the in-vehicle device 100-4 mounted on the vehicle D has a new feature candidate that matches the I-caching signboard template and the E-gas station signboard template indicated by the third feature parameter. A determination result indicating that there is no data is transmitted to the center 200. In addition, the transmission unit 102c of the vehicle-mounted device 100-5 mounted on the vehicle E matches the new feature candidate with the E gas station signboard template and the T car showroom signboard template indicated by the third characteristic parameter. A determination result indicating that it has not been transmitted is transmitted to the center 200.

  Whether the reception unit 202b of the center 200 matches the feature point indicated by the feature parameter of the next layer from the in-vehicle device 100 mounted on each of the plurality of vehicles to which the feature parameter of the next layer is transmitted in Step SB-14. The determination result for determining whether or not is received (step SB-19). For example, in step SB-19, the reception unit 202b transmits the same red rectangle new transmitted in step SB-18 by the processing of the transmission unit 102c of the in-vehicle devices 100-3 to 5 mounted in the vehicles C to E, respectively. A plurality of determination results related to feature candidates are received.

  The update unit 202c of the center 200 starts from the common item of the next layer based on the plurality of determination results received from the in-vehicle device 100 mounted on each of the plurality of vehicles by the processing of the reception unit 202b in Step SB-19. A feature point of a candidate for a new feature is determined (step SB-20). For example, in step SB-20, the update unit 202c transmits the vehicle C transmitted in step SB-18 by the processing of the transmission unit 102c of the in-vehicle devices 100-3 to 5 mounted on the plurality of vehicles C to E, respectively. Based on the determination result, the determination result of the vehicle D, and the determination result of the vehicle E, the feature point of the candidate for the new feature is determined from the common item of the next hierarchy. Specifically, the determination result of the vehicle C is that the new feature candidate matches the template of the M hamburger sign indicated by the third characteristic parameter, and the determination result of the vehicle D and the determination result of the vehicle E are Since the candidate of the object does not match the template of the sign of the I caching, E gas station, and T car showroom indicated by the third feature parameter, the feature point regarding the candidate template of the new feature is represented by M Decide as a hamburger sign.

  The update unit 202c of the center 200 determines whether or not the new feature can be specified based on the feature points of the new feature candidates determined in Step SB-13 and Step SB-20 (Step SB-21). . For example, in step SB-21, the update unit 202c identifies the new feature by referring to the feature parameter database 206b based on the feature points related to the color, shape, and template of the new feature candidate determined so far. Determine whether it is possible. Specifically, the updating unit 202c determines that the template similar to the M hamburger signboard template determined in step SB-20 does not exist in the common item arranged in the third hierarchy, and can be identified. When a similar template exists in the common item arranged in the third hierarchy, it is determined that it cannot be specified.

  If the update unit 202c of the center 200 determines in step SB-21 that a new feature can be specified (step SB-21: Yes), the update unit 202c proceeds to the next step SB-22. On the other hand, if the update unit 202c of the center 200 determines in step SB-21 that the new feature cannot be specified (step SB-21: No), the update unit 202c returns to the process of step SB-14 and again uses a similar template. Steps SB-14 to 21 are repeated.

  Hereinafter, a case will be described in which the common term arranged in the third hierarchy shifts to the process of step SB-22 on the assumption that a template similar to the M hamburger does not exist.

  When the update unit 202c of the center 200 can identify a new feature based on the feature points of the candidate for the new feature determined in Step SB-20 (Step SB-21: Yes), the determined new feature A new feature is identified based on the candidate feature points (step SB-22). For example, in step SB-22, the update unit 202c specifies that the new feature is an M hamburger sign based on the template for the M hamburger sign determined as the feature point of the candidate for the new feature.

  The update unit 202c of the center 200 updates the feature information stored in the feature information database 206a by associating the new feature specified in step SB-22 with the position information of the new feature (step SB). -23). For example, in step SB-23, the update unit 202c associates the M hamburger with the position information of the M hamburger, and updates the feature information stored in the feature information database 206a. Thereafter, the center 200 ends the process.

  Thus, in the feature information collection system according to the present embodiment, when there are new feature candidates newly established on the road, the center 200 detects a plurality of vehicles that can detect the new feature candidates. Thus, different feature parameters are transmitted from the common item of the feature parameter hierarchy stored in the feature parameter database 206b. Then, the vehicle detects a new feature candidate by the detection unit 110, and whether or not the detected new feature candidate matches the feature point of the common item in the hierarchy indicated by the feature parameter received from the center 200. And the determination result is transmitted to the center 200. Then, the center 200 determines the feature points of the new feature candidates from the common items of the hierarchy based on the determination results received from the plurality of vehicles. A determination process for determining a feature point of a candidate for a new feature executed by the center 200 and a plurality of vehicles is repeated for each hierarchy. Then, the center 200 identifies the new feature based on the feature points of the candidate for the new feature determined corresponding to each layer by this determination process, and the identified new feature and the position information of the new feature And the feature information stored in the feature information database 206a is updated. At this time, the center 200 determines the feature point of the candidate for the new feature from the common item of the hierarchy by the determination process, and then selects from the common item of the next layer associated with the determined feature point of the candidate for the new feature The different characteristic parameters are transmitted to a plurality of vehicles capable of detecting the candidate for the new feature, so that the determination process is executed in the next hierarchy.

  Thereby, when the feature information collection system according to the present embodiment detects a candidate for a new feature, by dividing the process of specifying the new feature according to the hierarchically arranged feature parameters into a plurality of stages, A process of gradually narrowing down new feature candidates can be executed. As a result, a process for identifying new features can be shared by a larger number of vehicles. In addition, since there are an infinite number of templates, it is possible to accurately identify a new feature while reducing the processing load on one vehicle without executing a vehicle other than the one that requires template matching.

  Here, in order to specify a new feature with higher accuracy, as shown in FIGS. 6 to 14 below, after confirming feature points of new feature candidates determined in each hierarchy with a plurality of vehicles. It is desirable to execute a process of proceeding to the next hierarchy and gradually narrowing down new feature candidates. Hereinafter, the process in this case will be described with reference to FIGS.

  First, with reference to FIGS. 6-12, an example of the operation | movement until it specifies the new feature performed with the center 200 concerning this embodiment and a some vehicle is demonstrated.

  As shown in FIG. 6, the transmission unit 202a of the center 200 recognizes a predetermined upper left range of the recognition region of the detection unit 110-1 for a plurality of vehicles including the vehicle A traveling on the road. The designation information and a first feature parameter for instructing to recognize a red object as a new feature candidate are transmitted. And the image analysis part 102b of the vehicle equipment 100-1 mounted in the vehicle A is the upper left among the images acquired from the detection part 110-1 according to the recognition range designation | designated information received from the center 200, and a 1st characteristic parameter. An image within the recognition range is analyzed to detect a red object installed in the traveling direction of the traveling road.

  As illustrated in FIG. 7, the transmission unit 102 c of the in-vehicle device 100-1 mounted on the vehicle A transmits a determination result indicating that the red rectangular object has been recognized to the center 200.

  As shown in FIG. 8, the transmitting unit 202a of the center 200 next checks the red rectangular object with respect to the vehicle B following the vehicle A in order to check whether the red rectangular object recognized by the vehicle A exists. The first feature parameter and the second feature parameter for instructing to recognize the are transmitted. And the image analysis part 102b of the vehicle equipment 100-2 mounted in the vehicle B is an upper left among the images acquired from the detection part 110-2 according to the 1st characteristic parameter and 2nd characteristic parameter which were received from the center 200. The image within the recognition range is analyzed to detect a red rectangular object installed in the traveling direction of the road that is running. And the transmission part 102c of the vehicle equipment 100-2 mounted in the vehicle B transmits the determination result which shows having recognized the red rectangular object to the center 200 similarly to the vehicle A. FIG.

  As shown in FIG. 9, the transmission unit 202a of the center 200 next identifies some specific objects for the vehicle C following the vehicle B in order to identify the red rectangular object recognized by the vehicle A and the vehicle B. Is transmitted as a recognition target. Here, the transmission unit 202a transmits to the vehicle C the third characteristic parameter indicating the I-caching signboard template and the E-gas station signboard template. And the image analysis part 102b of the vehicle equipment 100-3 mounted in the vehicle C is an image in the upper left recognition range among the images acquired from the detection part 110-3 according to the 3rd characteristic parameter received from the center 200. Perform pattern matching for. Here, the image analysis unit 102b performs pattern matching on an image including a red rectangular object that is a candidate for a new feature by using a template for an I-caching signboard and a signboard template for an E gas station. By this pattern matching, the image analysis unit 102b obtains a determination result that the red rectangular object, which is a candidate for a new feature, does not match the I-caching signboard template or the E-gas station signboard template. The transmitting unit 102c of the in-vehicle device 100-3 mounted on the vehicle C confirms that the red rectangular object that is a candidate for a new feature does not match the template for the I-caching signboard or the signboard for the E-gas station. Is sent to the center 200.

  As shown in FIG. 10, the transmission unit 202a of the center 200 receives the content of the determination result received from the vehicle C, and then receives a third feature parameter including a template different from the third feature parameter transmitted to the vehicle C. select. Then, the transmission unit 202a transmits the third feature parameter indicating the selected M hamburger sign template and the I-caching sign template to the vehicle D following the vehicle C. And the image analysis part 102b of the vehicle equipment 100-4 mounted in the vehicle D according to the 3rd characteristic parameter received from the center 200 WHEREIN: The image in the upper left recognition range among the images acquired from the detection part 110-4. Perform pattern matching for. Here, the image analysis unit 102b performs pattern matching on an image including a red rectangular object that is a candidate for a new feature using an M hamburger signboard template and an I-caching signboard template. By this pattern matching, the image analysis unit 102b obtains a determination result that the red rectangular object that is a candidate for a new feature matches the template of the M hamburger sign but does not match the template of the I caching sign. Then, the transmitting unit 102c of the in-vehicle device 100-4 mounted on the vehicle D has a red rectangular object that is a candidate for a new feature matches the template of the M hamburger signboard, Transmits a determination result indicating that they do not match to the center 200.

  As shown in FIG. 11, the transmitting unit 202a of the center 200 is similar to the signboard template of the M hamburger in order to confirm the determination result that the red rectangular object transmitted from the vehicle D is the M hamburger. A third feature parameter including the M2 hamburger sign template is selected. Then, the transmission unit 202a transmits the third feature parameter indicating the selected M hamburger sign template and the M2 hamburger sign template to the vehicle E following the vehicle D. And the image analysis part 102b of the vehicle equipment 100-5 mounted in the vehicle E is an image in the upper left recognition range among the images acquired from the detection part 110-5 according to the 3rd characteristic parameter received from the center 200. Perform pattern matching for. Here, the image analysis unit 102b performs pattern matching on an image including a red rectangular object that is a candidate for a new feature by using an M hamburger signboard template and an M2 hamburger signboard template. By this pattern matching, the image analysis unit 102b obtains a determination result that the red rectangular object that is a candidate for a new feature matches the signboard template of the M hamburger, but does not match the template of the M2 hamburger signboard. And the transmission part 102c of the vehicle equipment 100-5 mounted in the vehicle E matches the signboard template of the M2 hamburger with the red rectangular object that is a candidate for the new feature matches the signboard template of the M2 hamburger. Transmits a determination result indicating that they do not match to the center 200. At this time, the transmission unit 102c also confirms the determination result of the vehicle D that the red rectangular object is the sign of the M hamburger even in the vehicle E, and thus the sign of the M hamburger sign calculated by the relative position calculation unit 102e. The position information is also transmitted to the center 200 together.

  As shown in FIG. 12, the update unit 202 c of the center 200 next determines the red rectangular object transmitted from the vehicle E based on the determination result that the M hamburger signboard and the position information of the M hamburger signboard. The feature information of the new feature is newly registered and updated in the feature information database 206a. And the transmission part 202a of the center 200 transmits the feature information which shows that M hamburger exists as a new landmark on the road where the vehicle F travels to the vehicle F following the vehicle E. Accordingly, the vehicle F can use the feature information of the new feature received from the center 200 for uses such as driving assistance and route guidance.

  Subsequently, the details of the processing of the feature information collection system corresponding to FIGS. 6 to 12 described above will be described with reference to FIGS. 13 to 14.

  As illustrated in FIG. 13, the transmission unit 102 c of the vehicle A inquires of the center 200 about a characteristic parameter for recognizing a new feature candidate (step SC-1). For example, in step SC-1, the transmission unit 102c of the vehicle A performs the timing when the engine of the vehicle is operated and the timing when the current position of the vehicle enters a predetermined area (for example, the timing when the vehicle enters the suburbs). The center 200 is inquired about the characteristic parameters.

  Next, the transmission unit 202a of the center 200 transmits, to the vehicle A, the first characteristic parameter assigned to the vehicle A determined together with the recognition range designation information to the vehicle A (step SC-2). Here, for example, the center 200 transmits the first feature parameter to the vehicle A so as to recognize the red object within the recognition range at the upper left.

  Next, the image analysis unit 102b of the vehicle A recognizes a red object by operating a recognition algorithm according to parameters assigned to the vehicle A (step SC-3). In this recognition algorithm, when a red object designated by the first feature parameter is recognized, a second feature parameter that designates a shape such as a rectangle, a diamond shape, a circle, or an ellipse associated with the red color of the first feature parameter. The shape determination along the line may be included. Then, the image analysis unit 102b of the vehicle A determines whether or not a red rectangular object is recognized by the recognition algorithm operated in step SC-3. The shape of the red rectangular object is determined along the second feature parameter (step SC-4). When it is determined in step SC-4 that the red rectangular object is recognized (step SC-4: Yes), the transmitting unit 102c of the vehicle A has recognized the red rectangular object within the upper left recognition range. Is sent to the center 200 (step SC-5). Here, in Step SC-4, when it is determined that the red rectangular object is not recognized (Step SC-4: No), the process returns to Step SC-3.

  Next, when the transmitting unit 202a of the center 200 receives a determination result indicating that the red rectangular object is recognized from the vehicle A by the receiving unit 202b, the transmitting unit 202a passes the point where the red rectangular object recognized by the vehicle A exists. By transmitting the first feature parameter and the second feature parameter for instructing other vehicles to recognize the red rectangular object within the recognition range in the upper left, the feature parameter is assigned (step SC- 6).

  Next, the image analysis unit 102b of each vehicle to which the characteristic parameter is assigned in step SC-6 determines the presence or absence of a red rectangular object by image analysis, and transmits the determination result to the center 200 by the transmission unit 102c (step step). SC-7).

  Next, the control unit 202 of the center 200 determines whether or not the red rectangular object recognized by the vehicle A within the upper left recognition range is surely present based on the determination result transmitted from each vehicle in step SC-7. Is determined (step SC-8). For example, in step SC-8, the control unit 202 determines that at least one of the determination results transmitted from each vehicle in step SC-7 has recognized a red rectangular object within the upper left recognition range. If it is, it is determined that a red rectangular object exists. On the other hand, when all of the determination results transmitted from the respective vehicles are determination results indicating that a red rectangular object has not been recognized within the upper left recognition range, the control unit 202 does not have a red rectangular object. Determine that there is a possibility.

  Next, when it is determined in step SC-8 that the red rectangular object is surely present (step SC-8: Yes), the transmission unit 202a of the center 200 determines the third feature parameter indicating the feature template, A feature parameter is assigned by transmitting a red rectangular object to each subsequent vehicle capable of being detected (step SC-10). For example, in step SC-10, the transmission unit 202a assigns a third characteristic parameter indicating a template for an M hamburger signboard to a certain vehicle, and a signboard template for an E gas station to another vehicle. Is assigned a third feature parameter. And it transfers to the process of step SC-11 shown in FIG.

  Here, when it is determined in step SC-8 that there is a possibility that the red rectangular object does not exist (step SC-8: No), the control unit 202 of the center 200 is newly recognized as a red rectangular object. The feature parameter of each succeeding vehicle traveling in front of the point where the candidate for feature exists is reset (step SC-9). In other words, the originally allocated parameters are reset. Thereafter, the process returns to step SC-3.

  As shown in FIG. 14, the image analysis unit 102b of each vehicle performs template matching using the parameters assigned in step SC-10 (step SC-11). For example, in step SC-11, the image analysis unit 102b of a certain vehicle performs template matching using the signboard template of the M hamburger, and the image analysis unit 102b of another vehicle uses the signboard template of the E gas station. Template matching.

  Next, the control unit 102 of each vehicle recognizes by template matching using the template indicated by the assigned parameter (in FIG. 14, the M hamburger template indicated by the assigned third characteristic parameter is referred to as a pattern Z). It is determined whether or not it has succeeded (step SC-12). For example, in step SC-12, the control unit 102 determines that the recognition is successful when the pattern Z allocated to each vehicle matches the red rectangular object. On the other hand, when the pattern Z allocated to each vehicle does not match the red rectangular object, the control unit 102 determines that the recognition has failed.

  Next, the control unit 202 of the center 200 determines that the recognition by the template matching has failed using the pattern Z by the process of the control unit 102 of the vehicle (Step SC-12: No), the center 200 from each vehicle. It is determined whether or not the recognition success information indicating that the recognition of the pattern Z transmitted to is successful has been received by the receiving unit 202b for a predetermined number (step SC-13).

  Here, if the control unit 202 of the center 200 has received a predetermined number of pieces of recognition success information based on the pattern Z in Step SC-13 (Step SC-13: No), the process proceeds to Step SC-15. . On the other hand, when the control unit 202 of the center 200 cannot receive the recognition success information for the predetermined number in step SC-13 (step SC-13: Yes), the new feature recognized as a red rectangular object. The feature parameter of each subsequent vehicle that travels in front of the point where the candidate is present is reset (step SC-14). In other words, the originally allocated parameters are reset. Thereafter, the process returns to step SC-3 in FIG.

  Returning to the process of step SC-12, when it is determined that the recognition by the template matching using the pattern Z is successful by the process of the control unit 102 (step SC-12: Yes), the transmission unit 102c of the vehicle has a red color. A determination result indicating that the rectangular object matches the pattern Z is transmitted to the center 200 (step SC-15). It is assumed that the vehicle resets the characteristic parameter and restarts the recognition algorithm after passing through a point where a red rectangular object recognized to match the pattern Z exists.

  Next, the transmitting unit 202a of the center 200 determines whether or not the red rectangular object matches the pattern Z for other vehicles that pass through the point where the red rectangular object recognized to match the pattern Z exists. In order to instruct to recognize these, the same third feature parameter is transmitted to assign the feature parameter (step SC-16).

  Next, the image analysis unit 102b of each vehicle to which the characteristic parameter is assigned in step SC-16 determines the presence / absence of a red rectangular object that matches the pattern Z by pattern matching, and the transmission unit 102c determines the determination result. It transmits to the center 200 (step SC-17).

  Next, the control unit 202 of the center 200 determines whether or not the red rectangular object recognized in the upper left recognition range matches the pattern Z based on the determination result transmitted from each vehicle in Step SC-17. Is determined (step SC-18). For example, in step SC-18, the control unit 202 determines that at least one of the determination results transmitted from each vehicle in step SC-17 matches the red rectangular object recognized within the upper left recognition range with the pattern Z. If the result of the determination indicates that the red rectangular object matches the pattern Z, it is determined. On the other hand, when all of the determination results transmitted from each vehicle in step SC-17 are determination results indicating that the red rectangular object recognized in the upper left recognition range does not match the pattern Z, the control unit 202 , It is determined that the red rectangular object may not match the pattern Z.

  Here, if it is determined in step SC-18 that the red rectangular object may not match the pattern Z (step SC-18: No), the control unit 202 of the center 200 recognizes it as a red rectangular object. The feature parameter of each subsequent vehicle that travels before the point where the candidate for the new feature that has been made exists is reset (step SC-19). In other words, the originally allocated parameters are reset. Thereafter, the process returns to step SC-3 in FIG.

  On the other hand, if the transmitting unit 202a of the center 200 determines in step SC-18 that the red rectangular object surely matches the pattern Z (step SC-18: Yes), the transmitting unit 202a of the center 200 can detect the red rectangular object. Recognize the new feature (the sign of M hamburger in FIG. 14) discovered in the process up to step SC-18 for the following vehicle, obtain the relative distance to the new feature, and obtain the GPS information of the vehicle. And a signal instructing to calculate the existence position of the new feature is transmitted (step SC-20).

  Next, when the relative position calculation unit 102e of each vehicle receives the signal transmitted from the center 200 in step SC-20, the relative position calculation unit 102e calculates the current position of the vehicle based on the current position information of the vehicle acquired by the vehicle position acquisition unit 102d. The relative position of the new feature relative to the current position of the vehicle is calculated based on the distance to the new feature (in FIG. 14, the sign of M hamburger) measured by the detection unit 110 such as a radar. Get the location information of an object. And the transmission part 102c of each vehicle transmits the positional information on the new feature acquired by the process of the relative position calculation part 102e to the center 200 (step SC-21).

  Next, the control part 202 of the center 200 acquires the exact position of the discovered new feature by statistically processing the position information of the new feature acquired from many vehicles (step SC-22).

  Next, the updating unit 202c of the center 200 updates the feature information by associating the exact position of the new feature acquired in step SC-22 with the new feature and newly registering it in the feature information database 206a. To do. Then, the transmission unit 202a of the center 200 uses the new feature so that it can be used as a landmark for driving assistance, route guidance, and the like for each subsequent vehicle that is traveling in front of the new feature installation point. The feature information of the feature is transmitted (step SC-23). Thereafter, this process is terminated.

  As described above, in the feature information collection system according to the present embodiment, the center 200 determines the feature points of the new feature candidate from the common item of the hierarchy by the determination process, and then determines the feature of the determined new feature candidate. The same feature parameter corresponding to the point is transmitted to a plurality of vehicles capable of detecting the candidate for the new feature. Then, the vehicle detects a candidate for a new feature by the detection unit 110, determines whether or not the detected feature candidate matches the feature point indicated by the feature parameter received from the center 200, The determination result is transmitted to the center 200. Then, based on the determination results received from the plurality of vehicles, the center 200 confirms whether the feature points of the new feature candidates determined from the common item of the hierarchy by the determination process are the correct results, and the confirmed new features Each of the different characteristic parameters selected from the common item of the next hierarchy associated with the candidate feature point is transmitted to a plurality of vehicles capable of detecting the candidate for the new feature.

  As a result, the feature information collection system according to the present embodiment confirms the feature points of the new feature candidates determined in each hierarchy with a plurality of vehicles, and then proceeds to the next hierarchy and gradually finds new feature candidates. Therefore, it is possible to specify a new feature with higher accuracy while reducing the processing load per vehicle.

  In the above embodiment, when the center 200 allocates the feature parameters stored in the feature parameter database 206b to a plurality of vehicles, as shown in FIG. Although the case where the characteristic parameters are randomly assigned to capture the red object for the vehicle A, the green object for the vehicle B, and the blue object for the vehicle C has been described as an example, the present invention is not limited to this. In FIG. 15, the DB including the assumed matching template is a red object such as a stop sign, an entry prohibition sign, a red signal, an M hamburger, an I caching, an E gas station, a T car showroom, an N car dealer, etc. The template is remembered. In addition, a guide sign, a green light, an M2 hamburger, a J gas station, a T car store, etc. are stored as green objects.

  In the above embodiment, as shown in FIG. 16, the center 200 may determine the characteristic parameter to be allocated by reflecting the personal preference of the user of each vehicle from the DB including the assumed matching template. . The center 200 may assign a characteristic parameter for capturing a red object to the vehicle A of a user who likes the M hamburger so as to extract the M hamburger. Further, the center 200 may assign a feature parameter for capturing a green object to a vehicle B of a user who likes the M2 hamburger so as to extract the M2 hamburger. In this case, a new feature desired by the user of each vehicle is extracted, and the number of landmarks according to the user's preference can be increased.

  In addition, the center 200 may determine a characteristic parameter to be allocated by reflecting the vehicle type of each vehicle. For example, characteristic parameters that extract roadside stations or convenience stores that can be rest areas may be assigned to large trucks. In addition, when the center 200 holds information such as the age or sex of the driver or the vehicle owner, the family structure, etc., the center 200 may allocate characteristic parameters according to the information of the driver or the vehicle owner. Further, the center 200 may change the characteristic parameter to be allocated according to the time zone. For example, it becomes difficult to recognize objects with dark colors or complex shapes or patterns at night, so feature parameters are assigned to extract objects with bright colors, and dark objects are extracted during the day. The ratio of allocating feature parameters may be increased.

100 vehicle-mounted device 102 control unit 102a reception unit 102b image analysis unit 102c transmission unit 102d vehicle position acquisition unit 102e relative position calculation unit 104 communication control interface unit 108 input / output control interface unit 112 position acquisition unit 114 display unit 116 input unit 118 audio output Unit 110 detection unit 200 center 202 control unit 202a transmission unit 202b reception unit 202c update unit 204 communication control interface unit 206 storage unit 206a feature information database 206b feature parameter database 206c recognition range designation information file 500 position transmission device

Claims (9)

A plurality of vehicles having at least a detection unit for detecting a feature installed in a traveling direction of the road;
A center having at least a storage unit that stores feature information in which the feature is associated with position information of the feature, and feature parameters corresponding to a plurality of feature points that specify the feature;
A feature information collecting system in which the plurality of vehicles and the center are communicably connected,
When there are new feature candidates newly established on the road, the center transmits different characteristic parameters to a plurality of vehicles capable of detecting the new feature candidates,
The vehicle detects the new feature candidate by the detection unit, and determines whether or not the detected new feature candidate matches the feature point indicated by the feature parameter received from the center. And send the determination result to the center,
The center determines feature points of the new feature candidate based on the determination results received from the plurality of vehicles, and identifies the new feature based on the determined feature point of the new feature candidate The feature information collection system is characterized in that the specified new feature and the position information of the new feature are associated with each other and the feature information stored in the storage unit is updated. The storage unit stores the feature parameters arranged hierarchically according to the common terms of the feature points,
When there is a new feature candidate newly established on the road, the center determines a hierarchy of the feature parameters stored in the storage unit for a plurality of vehicles capable of detecting the new feature candidate. Send different characteristic parameters from the common term,
The vehicle detects the candidate for the new feature by the detection unit, and matches the feature point of the common item of the hierarchy indicated by the feature parameter received from the center with respect to the detected candidate for the new feature Determine whether or not to send the determination result to the center,
The center repeats a determination process for determining a feature point of a candidate for a new feature from a common term of the hierarchy based on the determination result received from the plurality of vehicles for each hierarchy,
The new feature is identified based on the feature points of the candidate for the new feature determined corresponding to each hierarchy by the determination process, and the identified new feature and the position information of the new feature are obtained. The feature information collection system according to claim 1, wherein the feature information stored in the storage unit is updated in association with each other.   The center determines the feature point of the candidate for the new feature from the common item of the hierarchy by the determination process, and then starts from the common item of the next layer associated with the determined feature point of the candidate for the new feature. The selected different characteristic parameters are transmitted to a plurality of vehicles capable of detecting the candidate for the new feature, thereby causing the determination process to be executed in the next hierarchy. Feature information collection system. The center determines the feature point of the candidate for the new feature from the common item of the hierarchy by the determination process, and then sets the same feature parameter corresponding to the determined feature point of the candidate for the new feature to the new feature Send to multiple vehicles that can detect feature candidates,
The vehicle detects the new feature candidate by the detection unit, and determines whether or not the detected new feature candidate matches the feature point indicated by the feature parameter received from the center. And send the determination result to the center,
The center confirms based on the determination results received from the plurality of vehicles whether or not the feature points of the new feature candidates determined from the common item of the hierarchy by the determination process are correct results. Transmitting different feature parameters selected from the common items of the next hierarchy associated with the feature points of the new feature candidate to a plurality of vehicles capable of detecting the candidate of the new feature, The feature information collecting system according to claim 3.   5. The method according to claim 2, wherein each of the layers includes any one of a color of the feature, a shape of the feature, and a template of the feature as a common term. Feature information collection system. The center includes recognition range designation information for designating a recognition range in which the detection unit recognizes the candidate for the new feature from a recognition region that can be recognized by the detection unit for a plurality of vehicles traveling on the road. Send
When the detection unit detects the new feature candidate within the recognition range designated by the recognition range designation information received from the center, the vehicle indicates that the new feature candidate exists. Send presence information to the center,
The ground according to any one of claims 1 to 5, wherein when the presence information is received from the vehicle, the center determines that there is a candidate for a new feature newly established on the road. Material information collection system. A plurality of vehicles having at least a detection unit that detects a feature installed in a traveling direction of a road, and is communicably connected to the feature information in which the feature is associated with position information of the feature; and A center having at least a storage unit for storing feature parameters corresponding to a plurality of feature points specifying the feature,
When there are new feature candidates newly established on the road, the center transmits different characteristic parameters to the plurality of vehicles capable of detecting the new feature candidates, and from the plurality of vehicles. The feature point of the candidate for the new feature is determined based on the received determination result, the new feature is identified based on the feature point of the candidate for the new feature determined, and the identified new feature and the In association with the position information of the new feature, the feature information stored in the storage unit is updated,
The determination result matches the feature point indicated by the feature parameter received from the center with respect to the detected new feature candidate in the vehicle. A center characterized in that it is determined whether or not to transmit the result to the center. A plurality of vehicles having at least a detection unit for detecting a feature installed in a traveling direction of the road;
A center having at least a storage unit that stores feature information in which the feature is associated with position information of the feature, and feature parameters corresponding to a plurality of feature points that specify the feature;
A feature information collection method executed in a feature information collection system in which the plurality of vehicles and the center are connected to be communicable,
In the center, when there are new feature candidates newly established on the road, different characteristic parameters are transmitted to a plurality of vehicles capable of detecting the new feature candidates,
In the vehicle, the new feature candidate is detected by the detection unit, and whether or not the detected new feature candidate matches the feature point indicated by the feature parameter received from the center. Is determined, the determination result is transmitted to the center,
At the center, feature points of the candidate for the new feature are determined based on the determination results received from the plurality of vehicles, and the feature of the candidate for the new feature is determined based on the determined feature point of the candidate for the new feature. A feature information collecting method characterized in that the feature information stored in the storage unit is updated by associating the identified new feature with the position information of the new feature. . A plurality of vehicles having at least a detection unit that detects a feature installed in a traveling direction of a road, and is communicably connected to the feature information in which the feature is associated with position information of the feature; and A feature information collection method executed at a center having at least a storage unit that stores feature parameters corresponding to a plurality of feature points that specify a feature,
In the center, when there are new feature candidates newly established on the road, different characteristic parameters are transmitted to a plurality of vehicles capable of detecting the new feature candidates, and the plurality of vehicles The feature point of the candidate for the new feature is determined based on the received determination result, the new feature is specified based on the determined feature point of the candidate for the new feature, and the specified new feature And the feature information stored in the storage unit is updated in association with the position information of the new feature,
The determination result matches the feature point indicated by the feature parameter received from the center with respect to the detected new feature candidate in the vehicle. It is determined whether or not to perform, and the feature information collecting method is a result transmitted to the center.

Images