JP5286952B2 - Moving purpose estimation apparatus and moving purpose estimation method - Google Patents

Description

  The present invention relates to a movement purpose estimation apparatus and a movement purpose estimation method for estimating a movement purpose and a movement destination of a vehicle.

Conventionally, there has been proposed a navigation device that estimates a destination based on information such as the position of the host vehicle and destination information stored in a predetermined travel information storage unit without an operator inputting the destination ( For example, see Patent Document 1.)
JP 2002-328035 A

  Since the technique described in Patent Document 1 described above estimates the destination using the past travel history, the frequency of the past travel history increases for behaviors that are addictive to the user of the host vehicle. Therefore, although it is possible to accurately estimate the destination, there is a problem that it is difficult to estimate a destination of a less addictive action such as going to a wedding.

  Therefore, the present invention has been proposed in view of the above-described circumstances, and provides a movement purpose estimation device and a movement purpose estimation method that can accurately estimate a destination of a less addictive action. Objective.

In the movement purpose estimation device and the movement purpose estimation method according to the present invention, the captured image, the decoration state of the vehicle occupant recognized from the image, and the movement purpose at that time are associated with each other and stored in the database. Feature information including clothes and ornaments as a person's decoration state while comparing the image and the decoration state recognized from the captured image, estimating the first movement purpose of the vehicle based on the comparison result And a travel purpose corresponding to the feature information, the second travel purpose of the vehicle based on the feature information stored in the database created using the information registered in the general-purpose network Is estimated.
The movement purpose estimation device and the movement purpose estimation method according to the present invention preferentially estimate the first movement purpose when the first movement purpose and the second movement purpose are not significantly different from each other. If the second movement purpose is significantly different from the second moving purpose, the decoration state recognized from the captured image is a special condition, the captured image, the decoration state of the vehicle occupant recognized from the image, and the current state The above-mentioned problem is solved by associating the purpose of movement with the database and recording it in the database .

  According to the movement purpose estimation device and the movement purpose estimation method according to the present invention, since the decoration state of the vehicle occupant is recognized in order to obtain the movement purpose, it is possible to accurately estimate the destination of the less addictive action. it can.

  Hereinafter, the movement purpose estimation apparatus as a preferred embodiment of the present invention will be described in detail.

[First Embodiment]
[Configuration of moving purpose estimation device]
As shown in FIG. 1, the movement purpose estimation apparatus shown as the first embodiment of the present invention stores a map database 11 that stores map data, a traffic information database 12 that stores traffic information data, and web information data. Web information database 13, broadcast database 14 for storing broadcast data, support information database 15 for storing support information data useful for the driver of the own vehicle, and a passenger of the own vehicle created based on general information Decoration state / movement purpose database 16 (second storage means) that stores the decoration state, movement purpose, and movement destination as data, an in-vehicle camera 21 that captures the situation in the host vehicle, and a CAN (Controller Area Network) ) To determine the position of the host vehicle, and a CAN signal acquisition unit 22 that acquires a CAN signal including vehicle data acquired via A host vehicle position determination unit 31, a road information acquisition unit 32 that acquires information about a road on which the host vehicle travels, a video analysis recognition unit 33 that analyzes and recognizes video information captured by the in-vehicle camera 21, and a host vehicle A decoration state / location information acquisition unit 41 that acquires a decoration state, a movement purpose, and a movement destination of an occupant riding in the vehicle; a movement location determination unit 42 that determines a location where the host vehicle has reached the movement destination; The decoration state / movement location storage unit 43 (first storage means) that stores the decoration state, movement purpose, and movement destination of the occupant of the own vehicle created based on the personal information as data, and the state of the own vehicle The vehicle state determination unit 44 for determining, the movement purpose estimation unit 45 for estimating the current movement purpose and the movement destination of the own vehicle, and the movement purpose and the movement destination estimated by the movement purpose estimation unit 45 Presentation A presentation information selection unit 46 for selecting information, a destination setting unit 51 for setting a destination for route guidance of the host vehicle, and presentation information selected by the presentation information selection unit 46 are presented to the passenger of the host vehicle. And an information presentation unit 52.

  The map database 11 is composed of, for example, a hard disk or the like, and stores map data including node information and link information representing road information and intersection information used for route guidance of the host vehicle. The map data also stores facility information such as store information attached to node information and link information. The map data stored in the map database 11 is read by the own vehicle position determination unit 31 and the road information acquisition unit 32.

  The traffic information database 12 includes, for example, a hard disk and stores traffic information data such as VICS information data that provides various information such as traffic congestion on a road by FM multiplex broadcasting or the like. The traffic information data stored in the traffic information database 12 is read by the road information acquisition unit 32.

  The web information database 13 includes, for example, a hard disk, and stores web information data obtained by accessing the Internet, which is a general-purpose network, as well as video captured by a web camera. The web information data stored in the web information database 13 is read by the decoration state / location information acquisition unit 41 and the presentation information selection unit 46.

  The broadcast database 14 is composed of, for example, a hard disk or the like, and stores broadcast data received by a receiver that receives a broadcast service such as terrestrial digital broadcasting that can exchange information between both banks. Broadcast data stored in the broadcast database 14 is read by the decoration state / location information acquisition unit 41 and the presentation information selection unit 46.

  The support information database 15 is composed of, for example, a hard disk and stores support information data useful for the driver of the host vehicle. Specifically, the support information data is data obtained from, for example, a service that distributes information for supporting route guidance of the host vehicle from a predetermined information distribution center, other dedicated information communication services, and the Internet. Data received via communication means such as the telephone 18. The support information data stored in the support information database 15 is read by the decoration state / location information acquisition unit 41 and the presentation information selection unit 46.

  The decoration state / movement purpose database 16 includes, for example, a hard disk or the like, and based on the decoration state / location information acquired by the decoration state / location information acquisition unit 41, the movement purpose and movement corresponding to the decoration state of the occupant of the own vehicle. The destination is stored as data. That is, the decoration state / movement purpose database 16 is based on general information such as web information data read from the web information database 13, broadcast data read from the broadcast database 14, and support information data read from the support information database 15. Created. The decoration state / movement purpose database 16 may be stored in a hard disk or the like in the host vehicle, or may be stored in a predetermined information distribution center that distributes information. The data stored in the decoration state / movement purpose database 16 is read out by the movement purpose estimation unit 45.

  The GPS signal receiving unit 17 acquires GPS information necessary for specifying the position of the host vehicle based on the GPS signal distributed from the GPS satellite. The GPS information acquired by the GPS signal receiving unit 17 is supplied to the own vehicle position determining unit 31.

  The in-vehicle camera 21 has a position and an imaging range for capturing an image of a situation in the host vehicle such as a decoration state or belongings of a passenger riding in the host vehicle. Video information captured by the in-vehicle camera 21 is supplied to the video analysis recognition unit 33.

  The CAN signal acquisition unit 22 is connected to a CAN communication line mounted on the vehicle. The CAN communication line is connected to sensors for acquiring vehicle speed, engine speed, shift position, steering angle, accelerator pedal opening, brake pedal operation, turn signal operation, and the like as vehicle data. And the CAN signal acquisition part 22 acquires the CAN signal for communicating the vehicle data regarding the own vehicle collectively. Specifically, the CAN signal output unit 22 acquires vehicle data such as a vehicle speed, an engine speed, a shift position, a steering angle, an accelerator pedal opening, a brake pedal operation, and a winker operation as a CAN signal. The CAN signal acquisition unit 22 supplies the acquired CAN signal to the own vehicle state determination unit 44.

  The own vehicle position determining unit 31 determines where the own vehicle position is on the map based on the GPS information acquired by the GPS signal receiving unit 17 and the map data read from the map database 11. The own vehicle position determination unit 31 supplies the determined own vehicle position information to the movement location determination unit 42.

  The road information acquisition unit 32 acquires road information related to the road on which the vehicle travels based on the map data read from the map database 11 and the traffic information data read from the traffic information database 12. The road information acquisition unit 32 supplies the acquired road information to the movement location determination unit 42.

  The video analysis recognizing unit 33 analyzes video information captured by the in-vehicle camera 21 and recognizes a decoration state or the like of an occupant on the own vehicle. The video analysis recognition unit 33 supplies recognition information indicating the recognition result to the movement purpose estimation unit 45.

  The decoration state / location information acquisition unit 41 reads the web information data read from the web information database 13, the broadcast data read from the broadcast database 14, and the support information database 15 to create the decoration state / movement purpose database 16. Based on the support information data and the like, the decoration state, the movement purpose, and the movement destination of the occupant on the own vehicle are acquired. The decoration state / location information acquisition unit 41 stores the decoration state / location information indicating the acquired decoration state, movement purpose, and movement destination of the occupant in the decoration state / movement purpose database 16. In this embodiment, the case where both the movement purpose and the movement destination that matches the movement purpose can be acquired will be described. However, only the movement purpose may be acquired from the decoration state of the occupant.

  The moving place determination unit 42 includes the own vehicle position information supplied from the own vehicle position determination unit 31, the road information supplied from the road information acquisition unit 32, the own vehicle state information supplied from the own vehicle state determination unit 44, and the like. Based on the above, the location where the host vehicle has reached as the moving destination is determined. The movement location determination unit 42 stores the determined movement location information in the decoration state / movement location storage unit 43.

  The decoration state / movement location storage unit 43 is composed of, for example, a hard disk or the like, and is based on the recognition information supplied from the video analysis recognition unit 33 and the movement location information supplied from the movement location determination unit 42. The movement purpose and the movement destination corresponding to the decoration state are stored as data. That is, the decoration state / movement location storage unit 43 is different from the decoration state / movement purpose database 16, and includes personal information such as recognition information supplied from the video analysis recognition unit 33 and movement location information supplied from the movement location determination unit 42. That is, it is created based on information unique to the host vehicle. The data stored in the decoration state / movement location storage unit 43 is read out by the movement purpose estimation unit 45.

  The own vehicle state determination unit 44 estimates the purpose of movement based on the CAN signal made up of the vehicle data supplied from the CAN signal acquisition unit 22 and includes whether the own vehicle has arrived at the movement destination. Determine the state of the vehicle. The own vehicle state determination unit 44 supplies the determined own vehicle state information to the movement location determination unit 42 and the presentation information selection unit 46.

  The movement purpose estimation unit 45 compares the recognition information supplied from the video analysis recognition unit 33 with the data stored in the decoration state / movement purpose database 16 and the decoration state / movement location storage unit 43, and the comparison result. Based on the above, the current travel destination and the travel destination of the host vehicle are estimated from the locations where the host vehicle has reached as the travel destination in the past. The movement purpose estimation unit 45 supplies the estimated movement purpose information to the presentation information selection unit 46 and the destination setting unit 51.

  The presentation information selection unit 46 includes map data read from the map database 11, traffic information data read from the traffic information database 12, web information data read from the web information database 13, broadcast data read from the broadcast database 14, and support information database. 15 based on the support information data read from 15 and the vehicle state information supplied from the vehicle state determination unit 44, etc., and the presentation information corresponding to the movement purpose and the movement destination estimated by the movement purpose estimation unit 45 are selected. To do. The presentation information selection unit 46 supplies the selected presentation information to the information presentation unit 52.

  Based on the travel purpose information supplied from the travel purpose estimation unit 45, the destination setting unit 51 sets a destination for performing route guidance for the host vehicle automatically or after confirming with the driver in a proposal type. . When the destination is set by the destination setting unit 51, the host vehicle can start route guidance according to the destination.

  The information presentation unit 52 is configured using, for example, a TTS (Text to Speech) system in a car navigation system, a liquid crystal display screen that displays visual information, a speaker that outputs auditory information, and the like. The selected presentation information is presented mainly to the occupants who are in the vehicle as voice information.

[Operation of moving purpose estimation device]
Such a movement purpose estimation apparatus including each unit estimates a movement purpose according to the decoration state of a passenger on the host vehicle according to a series of procedures as shown in FIG.

  First, the movement purpose estimation apparatus is normally performed when the vehicle starts to travel with the operation of the ignition key of the vehicle as a trigger. As shown in FIG. 2, the moving purpose estimation device determines whether or not there is a passenger in the own vehicle by analyzing the video imaged by the in-vehicle camera 21 by the video analysis recognition unit 33 in step S <b> 1. .

  Here, when the occupant is not seated at the normal seat position, the movement purpose estimation device recognizes that there is no occupant in the host vehicle, and proceeds to step S2. In step S2, it is determined that the host vehicle does not move, and the previously estimated movement purpose is reset in order to estimate a new movement purpose. However, the previous movement purpose is used when estimating the next movement purpose. Therefore, the movement purpose estimation device stores the movement purpose in a storage means such as a memory, which will be described later, at the stage when the estimation result is obtained, and the difference from the actual movement destination, that is, the movement purpose Information indicating whether the estimation result is as estimated or not estimated is also stored.

  On the other hand, when it is recognized that there is an occupant in the own vehicle, in step S3, in order to confirm the occupant's decoration state, the in-vehicle image (video) captured mainly by the in-vehicle camera 21 with the person zoomed up. To get. The movement purpose estimation device can perform personal recognition by the video analysis recognition unit 33 based on the face image portion of such an in-vehicle image, and can acquire a decoration state for each occupant.

  In step S4, the image analysis recognition unit 33 recognizes the occupant's face, neck, arms, torso, legs, and the like, and then extracts clothing information. Specifically, the image analysis and recognition unit 33 includes material information such as the length and shape of the sleeve, the neck, the presence or absence of a tie, the pattern (including the logo mark), the color, the texture, etc. of the clothing worn by the occupant. Extract as much clothing information as possible. The movement purpose estimation device stores the clothing information recognized by the video analysis recognition unit 33 in the decoration state / movement location storage unit 43.

  Also, in step S5, the clothing information recognized by the video analysis recognition unit 33 by the movement purpose estimation unit 45 is used as general clothing information stored in the decoration state / movement purpose database 16 and the decoration state / movement. It compares with the clothing information memorize | stored in the location memory | storage part 43. FIG. As a result, the movement purpose estimation unit 45 extracts a movement purpose and a movement destination having a high degree of similarity, and then proceeds to step S10.

  The decoration state / movement purpose database 16 is created based on general information as described above, and the creation process will be described later. Also, the calculation of similarity by comparing the clothing information recognized by the video analysis recognition unit 33 with the clothing information stored in the decoration state / movement purpose database 16 and the decoration state / movement location storage unit 43 will be described later. It shall be.

  In step S6, the video analysis recognition unit 33 recognizes the state of the face and head, such as the occupant's hairstyle, makeup, and wrinkle state, and extracts state information. In other words, when the purpose of travel is to go to a wedding or the like, if it is a female occupant, the hairstyle tends to be up and the makeup tends to be dark. In addition, when the purpose of travel is a job interview or a business trip, there is a tendency to shave, but when the purpose of travel is outing on a daily basis, it tends to be a state of stubbornness. As described above, the face and head states such as the occupant's hairstyle, makeup, and wrinkle state are information that contributes to the estimation of the purpose of movement.

  The movement purpose estimation device stores the state information recognized by the video analysis recognition unit 33 in the decoration state / movement location storage unit 43 and is recognized by the video analysis recognition unit 33 by the movement purpose estimation unit 45 in step S7. The state information is compared with general state information stored in the decoration state / movement purpose database 16 and state information stored in the decoration state / movement location storage unit 43. As a result, the movement purpose estimation unit 45 extracts a movement purpose and a movement destination having a high degree of similarity, and then proceeds to step S10. Note that the calculation of similarity by comparing the state information recognized by the video analysis recognition unit 33 with the state information stored in the decoration state / movement purpose database 16 and the decoration state / movement location storage unit 43 will also be described later. It shall be.

  Furthermore, in step S8, the image analysis recognition unit 33 recognizes features including decorations worn by the occupant, such as accessories and glasses, and extracts decoration information. Specifically, the video analysis recognition unit 33 extracts ornament information indicating as many features of the ornament as possible, such as pearl necklaces, earrings, rings, high-grade glasses, tie pins, cufflinks, and the like. The movement purpose estimation device stores the ornament information recognized by the video analysis recognition unit 33 in the decoration state / movement location storage unit 43, and also recognizes by the video analysis recognition unit 33 by the movement purpose estimation unit 45 in step S9. The decorated ornament information is compared with the general ornament information stored in the ornament state / movement purpose database 16 and the ornament information stored in the ornament state / movement location storage unit 43. And the movement purpose estimation part 45 transfers a process to step S10, after extracting the movement objective and movement destination with high similarity. It should be noted that the similarity is also calculated by comparing the decoration information recognized by the video analysis recognition unit 33 with the decoration information stored in the decoration state / movement purpose database 16 and the decoration state / movement location storage unit 43. This will be described later.

  Next, in step S10, the movement purpose estimation device 45 compares the movement purpose estimated in each of steps S5, S7, and S9 and the degree of fitness for the movement destination by the movement purpose estimation unit 45. Find the moving purpose and destination with the highest fitness.

  In the next step S11, the movement purpose estimation device determines the movement purpose and the movement destination obtained in step S10 as the final estimation result. At this time, the movement purpose estimation unit 45 is used for ceremonial occasions such as weddings, funerals, business trips, dates, concerts, and other events such as commuting, shopping, and pick-up. Compared to all travel purposes and travel destinations, including both highly customary travel purposes and travel destinations, such as hospital visits, lessons, sports, vehicle maintenance, etc. Determine the ground. Specifically, the movement purpose estimation unit 45 determines that there are things determined based on company uniforms, suits, ties, jackets, shirts with collars, etc. as highly addictive movement purposes and movement destinations. Estimate that commuting to the company is for mobile purposes. Moreover, the movement purpose estimation part 45 estimates that a pick-up is a movement purpose when there exists what was determined based on pajamas, home wear, etc. as a highly addictive movement purpose and a movement destination. Furthermore, the movement purpose estimation unit 45 can estimate that the movement is for sports when the occupant wears a jersey, a windbreaker, a hat, or the like. At this time, the movement purpose estimation unit 45 can further improve the estimation accuracy of the movement purpose and the movement destination by taking into account information such as the weather and time zone at the time of determination.

  In step S12, the movement purpose estimation apparatus 45 determines whether the movement purpose and the movement destination determined in step S11 coincide with the current position. Here, the movement purpose estimation device shifts the processing to step S14 and step S16 when the movement purpose and the movement destination do not coincide with the current location. On the other hand, if the travel purpose and the travel destination match the current location, in step S13, the travel purpose estimation unit 45 estimates the return home as the travel purpose, or if the current travel purpose is a wedding, For example, the next movement purpose related to the current movement purpose is estimated, such as movement to the society or movement to the crematorium after the farewell ceremony if it is a funeral. This can be estimated based on data indicating the transition of the past moving destination. The data indicating the transition of such a moving destination has a transition probability value like a Bayesian network, and a place with a high probability is extracted as the first candidate.

  When the movement purpose estimation device estimates the movement purpose and the movement destination in this way, in step S14, the map data read from the map database 11, the traffic information data read from the traffic information database 12, and the web information database 13 are read. Based on the web information data, the broadcast data read from the broadcast database 14, the support information data read from the support information database 15, the own vehicle state information supplied from the own vehicle state determination unit 44, and the like. The presentation information suitable for the movement purpose and the movement destination estimated by (1) is selected. Then, in step S15, the movement purpose estimation device presents the presentation information selected in step S14 to the occupant riding in the host vehicle as voice information, graphic information, character information, and the like by the information presentation unit 52. .

  In step S16, the destination estimation unit 51 automatically sets the destination for the route guidance of the host vehicle, or asks the driver to obtain an agreement if the destination is obtained. Or change the setting to end the series of processing.

  The movement purpose estimation device can appropriately estimate the movement purpose according to the decoration state of the occupant riding the host vehicle according to such a series of procedures.

[Creation process of decoration status / moving purpose database]
Next, a process for creating the decoration state / movement purpose database 16 will be described.

  As described above, the decoration state / movement purpose database 16 is a database created based on generally available information. In order to store information on fashionable hairstyles, information from the Internet or the like is sequentially received. It is updated using. Specifically, the decoration state / movement purpose database 16 is created according to a series of procedures as shown in FIG.

  First, in creating the decoration state / movement purpose database 16, as shown in FIG. 3, when a video of a web camera on the Internet such as a video of a person entering or leaving an arbitrary facility is acquired in step S21, step S22 is performed. , The clothing state is extracted by analyzing using a predetermined device having the same capability as the video analysis recognition unit 33. Thereafter, in step S23 and step S24, the state information such as hairstyle and makeup and the decoration information are extracted. From the viewpoint of protecting personal information, it is desirable to recognize up to gender without performing personal recognition. In addition, a method of preventing information outflow or the like by applying these processes collectively on a predetermined information distribution center side can also be applied.

  Further, in creating the decoration state / movement purpose database 16, in step S25, information on the decoration catalog on the Internet is acquired. In steps S26 to S28, information indicating the purpose of use of clothes and target customers, information indicating price, material, color, pattern, size, and the like are acquired, and the sales area, the number of sales, and the sales start date are acquired. Etc. are also acquired.

  Furthermore, in creating the decoration state / moving purpose database 16, in steps S29 to S32, in the same manner as the information extraction from the video of the webcam, the video at each location is used by using the video such as digital terrestrial broadcasting. Information on the decoration state of ordinary people is obtained, catalog information on television shopping, etc. is obtained.

  In step S33, the movement purpose estimation apparatus checks whether the information acquired in steps S21 to S32 is already registered in the decoration state / movement purpose database 16.

  Here, if the movement purpose estimation device is not registered in the decoration state / movement purpose database 16, in step S34 to step S36, for example, as shown in FIG. 4, the location (position) and decoration state (combination, etc.) In addition, the purpose of movement and the like in the case of being in the place are registered in association with each other, and the creation / update processing of the decoration state / movement purpose database 16 is ended.

  On the other hand, when the movement purpose estimation device has already been registered in the decoration state / movement purpose database 16, in step S37, only the information indicating the location and the decoration state is compared among the registered information. In step S38, it is determined whether or not the place and the decoration state are greatly different as a result of the comparison. If it is determined that the compared information is not substantially different and is substantially the same, the movement purpose estimation device proceeds to acquisition of other information without additionally registering the information acquired in steps S21 to S32. And migrate. In addition, if the compared information is greatly different and the movement purpose estimation device determines that the information is different, in step S39, as a more detailed condition, the date and time including the day of the week or the morning / afternoon, the season, After giving detailed conditions such as weather, sex, age group, etc., in step S34 to step S36, the location, decoration state, movement purpose, etc. are associated and registered, and the decoration state / movement purpose database 16 is created. finish.

  The decoration state / movement purpose database 16 can be created based on general information according to such a series of procedures.

  The movement purpose estimation apparatus can estimate the movement purpose when wearing a general decoration state by using the decoration state / movement purpose database 16 as described above. That is, the movement purpose estimation device recognizes the current decoration state of the passenger of the host vehicle by using the decoration state / movement purpose database 16 described above, and is registered in the decoration state and the decoration state / movement purpose database 16. The purpose of movement can be estimated by comparing with the information that is present.

[Similarity calculation processing]
The moving purpose estimation device stores the decoration state recognized by the video analysis recognition unit 33 in step S5, step S7, and step S9 in FIG. 2, the decoration state / movement purpose database 16, and the decoration state / movement location storage unit 43. The degree of similarity is calculated by comparing with the stored information indicating the decoration state. For example, when the occupant wearing a white tie in a black suit is imaged by the in-vehicle camera 21, the movement purpose estimation device refers to the decoration state / movement purpose database 16 by the movement purpose estimation unit 45, It is checked whether information indicating a decoration state combining a suit and a white tie is stored. When such information is stored in the decoration state / movement purpose database 16, that is, when the information indicating the decoration state matches, the movement purpose estimation device extracts the feature of the decoration state. Can do. In addition, in the case of such clothes, the movement purpose estimation device is stored in the decoration state / movement purpose database 16 by the movement purpose estimation unit 45 and its shape, color, pattern, logo mark and other images. It is also possible to calculate the degree of similarity with an image and extract and determine the one with the closest degree of similarity.

  There is a method of comparing such similarity based on a distance between two images by linearly complementing two different images, such as a morphing technique which is an image processing technique. Such a method has the advantage that it is quantitative, easy to judge and convenient. The morphing process is performed by determining which one of the two images is closer to the image by determining which of the two images is closer to the frame based on the number of frames between the two images. In addition, as a method of extracting the feature of the decoration state from the image, for example, in the case of a wedding, if a man wears a white tie on a black suit and the white chief comes out of the chest pocket a little Of the decoration states registered in the state / movement purpose database 16 and recognized by the video analysis recognition unit 33, those corresponding to the information registered in the decoration state / movement purpose database 16 may be extracted as features. . Furthermore, the movement purpose estimation device can estimate a movement purpose to go to a party or a meal when there is an occupant wearing a recently sold expensive dress.

  Specifically, the movement purpose estimation device calculates the similarity using morphing according to a series of procedures as shown in FIG.

  First, as shown in FIG. 5, the movement purpose estimation unit 45 is classified by genre as shown in FIG. 4 in step S42 when a new decoration state is recognized by the video analysis recognition unit 33 in step S41. Referring to the decoration state / movement purpose database 16, it recognizes clothing shapes such as sleeves and collars. In step S43, the movement purpose estimation unit 45 performs morphing based on the images recognized in step S41 and the clothing shape images classified by genre recognized in step S42. It is determined whether the morphing distance is 0.5 or more. Here, when the morphing distance is 0.5 or more, the movement purpose estimation unit 45 shifts the processing to step S45. On the other hand, if the morphing distance is less than 0.5, in step S44, the movement purpose estimation unit 45 adds a new clothing genre to the decoration state / movement purpose database 16, and the process proceeds to step S45. To do.

  Subsequently, in step S45, the movement purpose estimation unit 45 recognizes the pattern and color of the decoration state recognized by the video analysis recognition unit 33, and the pattern and position. In step S46, the movement purpose estimation unit 45 performs morphing based on the image recognized by the video analysis recognition unit 33 and the pattern and color images classified by genre recognized in step S45. It is determined whether the morphing distance between genres is 0.5 or more. Here, when the morphing distance is 0.5 or more, the movement purpose estimation unit 45 proceeds to step S48. On the other hand, if the morphing distance is less than 0.5, the movement purpose estimation unit 45 adds a new pattern or color genre to the decoration state / movement purpose database 16 and proceeds to step S48 in step S47. To migrate.

  Subsequently, the movement purpose estimation unit 45 recognizes the decorative material (texture) recognized by the video analysis recognition unit 33 in step S48. In step S49, the movement purpose estimation unit 45 performs morphing based on the image recognized by the video analysis recognition unit 33 and the image of the material classified by genre recognized in step S48. It is determined whether or not the morphing distance is 0.5 or more. Here, when the morphing distance is 0.5 or more, the movement purpose estimation unit 45 proceeds to step S51. On the other hand, if the morphing distance is less than 0.5, the movement purpose estimation unit 45 adds a new material genre to the decoration state / movement purpose database 16 in step S50 and shifts the process to step S51. To do.

  Then, in step S51, the movement purpose estimation unit 45 extracts the decoration item in the decoration state recognized by the video analysis recognition unit 33. In step S52, the movement purpose estimation unit 45 performs morphing based on the image recognized by the video analysis recognition unit 33 and the ornament images classified by genre extracted in step S51. It is determined whether the morphing distance between them is 0.5 or more. Here, the movement purpose estimation unit 45 repeats the processing from Step S41 when the morphing distance is 0.5 or more. On the other hand, if the morphing distance is less than 0.5, the movement purpose estimation unit 45 adds a new decoration item genre to the decoration state / movement purpose database 16 in step S53 and performs the processing from step S41. repeat.

  The movement purpose estimation device can calculate the similarity using morphing according to such a series of procedures, and can expand and update the contents of the decoration state / movement purpose database 16 according to the result.

  Note that the movement purpose estimation device recognizes not only the decoration state / movement purpose database 16 but also information indicating the decoration state stored in the decoration state / movement location storage unit 43 which is a database dedicated to the own vehicle. By comparing with the decoration state recognized by the unit 33, the estimation accuracy of the movement purpose and the movement destination is improved.

  Similarly to the decoration state / movement purpose database 16 based on general information, the decoration state / movement location storage unit 43 includes clothing information, state information such as hairstyle and makeup, and ornament information representing features such as accessories, The moving location information determined by the moving location determination unit 42 is stored. When the video captured by the in-vehicle camera 21 is recognized by the video analysis recognition unit 33, the movement purpose estimation device refers to the decoration state / movement location storage unit 43 by the movement purpose estimation unit 45 and is recognized by the video analysis recognition unit 33. The moving image and the information stored in the decoration state / movement location storage unit 43 are compared, and the movement purpose and the movement destination having the highest fitness are determined.

  At this time, the movement purpose estimation device uses the movement purpose and movement destination estimated in comparison with the decoration state / movement location storage unit 43 and the movement purpose and movement estimated in comparison with the decoration state / movement destination database 16. Unless the destination is significantly different, the moving destination and the moving destination based on the information stored in the decoration state / moving location storage unit 43 are preferentially estimated. On the other hand, the movement purpose estimation device includes the movement purpose and the movement destination estimated in comparison with the decoration state / movement place storage unit 43 and the movement purpose and movement purpose estimated in comparison with the decoration state / movement destination database 16. If it is significantly different from the ground, it is registered in the decoration state / movement location storage unit 43 as a unique condition, and then the suitability of the current result is determined, and a weight is given so that it can be fed back to the subsequent estimation.

  Thereby, the movement purpose estimation apparatus is a case where the movement destination determined based on the decoration state / movement purpose database 16 and the movement destination determined based on the decoration state / movement place storage unit 43 are the same. However, it is effective when the purpose of movement is different. For example, the movement purpose estimation device, when a facility where a sports facility and a shopping center are attached is determined as a movement destination, whether the purpose of movement is to perform sports or go shopping It can be estimated based on the decoration state. Therefore, when the purpose of travel is to go shopping, the travel purpose estimation device presents bargain information or when the purpose of travel is to travel, the tennis court is crowded or schooled. You can present the contents of the class. In this way, the movement purpose estimation device can present information suitable for each movement purpose, and is very effective when the priority of movement purposes is different even if the movement destination is the same.

  Note that the movement purpose estimation device may store the video captured by the in-vehicle camera 21 or the like without performing image recognition, and perform similarity determination using the image. May be. The same applies to a web camera or a broadcast video. Thereby, although the data size of a video | video becomes large, the movement purpose estimation apparatus can reduce that an estimation is mistaken by misrecognition of image recognition. Furthermore, the moving purpose estimation device performs both the analysis of the image without image recognition and the analysis based on the result of the image recognition, and preferentially uses the method having the good estimation accuracy result. May be. Furthermore, since the movement purpose estimation apparatus is expected to have different estimation accuracy depending on the genre or the like, both of these methods may be used for each genre. For example, the movement purpose estimation device uses the result of image recognition for the analysis using the decoration state / movement purpose database 16 based on general information such as a web camera, etc., and the decoration state and belongings of the occupant of the own vehicle May be analyzed in the form of video without image recognition.

  In addition, what is recognized from the image of the in-vehicle camera 21 or the like is not limited to the combination of the above-described ones, but may be separate from each other, or may be combined. Similarly, the movement purpose estimation device selectively uses either the decoration state / movement purpose database 16 and the decoration state / movement location storage unit 43 without estimating the movement purpose and the movement destination. It is also possible to use it for estimation.

  Furthermore, it is desirable that the movement purpose estimation device performs estimation by adopting a higher similarity degree or a lower estimation error when a conflicting situation occurs in the passenger's decoration state. At this time, if the similarity is the same, the movement purpose estimation device may estimate a movement purpose and a movement destination with a low occurrence frequency. Further, the movement purpose estimation device may perform weighting according to the priority in the passenger's family. Furthermore, the movement purpose estimation device, when there is a passenger, is expected that the purpose of the driver's movement will be lower than that of the passenger, so the driver's weight is lowered and the passenger's weight is reduced. You may make it estimate by making it high.

[Effect of the first embodiment]
As described above in detail, according to the movement purpose estimation device shown as the first embodiment of the present invention, the own vehicle is based on the decoration state of the occupant of the own vehicle recognized from the video of the situation in the own vehicle. The movement purpose and the movement destination of the subject vehicle can be estimated with high accuracy, and the destination of the less addictive behavior can be estimated with high accuracy. be able to.

  In addition, the movement purpose estimation device associates the image in the host vehicle with the movement purpose (ground) and stores them in the decoration state / movement location storage unit 43, and recognizes the decoration state recognized from the stored image and the captured image. Based on the comparison result, the current travel destination and travel destination of the host vehicle are estimated from the locations where the host vehicle has arrived as a travel destination in the past. The history with the moving destination can be created as an individual database, and the moving destination and the moving destination can be estimated with high accuracy based on this database.

  Further, the movement purpose estimation device associates the decoration state of the occupant of the own vehicle recognized from the captured image in the own vehicle with the movement purpose (ground) at that time in the decoration state / movement location storage unit 43. The stored decoration state and the decoration state recognized from the captured image are compared by the movement purpose estimation unit 45, and based on the comparison result, the host vehicle has arrived as a movement destination in the past. Since the current purpose and destination of the subject vehicle are estimated from the location, the decoration state can be stored as text data, creating a database that reduces the amount of information compared to storing video The movement purpose and the movement destination can be estimated with high accuracy based on this database.

  Furthermore, the movement purpose estimation device stores feature information indicating clothes and decorations as a person's decoration state and locations corresponding to these clothes and decorations in the decoration state / movement purpose database 16. Based on the memorized feature information indicating the clothes and decorations, the movement purpose and the movement destination of the own vehicle are estimated. For example, the clothes, the state of hairstyle, makeup, etc. The relationship with the movement destination can be grasped in detail, and the movement destination and the movement destination can be estimated with high accuracy.

  Moreover, according to this movement purpose estimation device, the general state of movement and the movement purpose database 16 are created and updated based on general information including information on the Internet, which is a general-purpose network. The moving destination can be estimated using the latest database.

  Furthermore, according to this movement purpose estimation device, the decoration state / movement purpose database 16 is created, extended, and updated based on general information including images of people in each place taken by a web camera on the Internet. Thus, it is possible to estimate a general movement destination and a movement destination using the latest database by using the video on the Internet as it is.

  Furthermore, according to the movement purpose estimation apparatus, either or both of the information stored in the decoration state / movement location storage unit 43 and the information stored in the decoration state / movement purpose database 16 are selected. Therefore, the movement purpose and the movement destination of the host vehicle are estimated, so that the movement purpose and the movement destination suitable for the situation can be estimated with high accuracy.

  Further, according to this movement purpose estimation device, the presentation information corresponding to the movement purpose and the movement destination estimated by the movement purpose estimation unit 45 is selected by the presentation information selection unit 46, and the selected presentation information is selected by the information presentation unit 52. By presenting to the passenger of the own vehicle, it is possible to actively present necessary and necessary information to the passenger from the estimation result of the moving purpose and the moving destination, and provide a more enjoyable driving environment.

  Furthermore, according to this movement purpose estimation apparatus, the destination setting unit 51 sets the movement place estimated by the movement purpose estimation unit 45 as the destination when performing route guidance of the host vehicle, thereby enabling the destination setting. It is possible to save the time and effort of the occupant who performs the operation.

  Furthermore, according to this travel purpose estimation apparatus, the destination setting unit 51 sets the travel destination estimated by the travel purpose estimation unit 45 as a destination when performing route guidance of the host vehicle. Since the travel destination is set as the destination after inquiring the passenger of the host vehicle, the destination can be set in accordance with the driver's intention while reducing troublesome work.

[Second Embodiment]
Next, a second embodiment of the present invention will be described.

  In the second embodiment, as a decoration state, an occupant's belongings and an article brought into the host vehicle are recognized by an imaging device such as an in-vehicle camera and an image processing device, and a movement purpose and a movement destination are estimated based on the result. To do. Therefore, in this 2nd Embodiment, about the site | part same as the site | part demonstrated in 1st Embodiment, respectively, the same code | symbol is attached | subjected and the detailed description is abbreviate | omitted.

[Configuration of moving purpose estimation device]
As shown in FIG. 6, the movement purpose estimation apparatus shown as the second embodiment of the present invention includes the map database 11, the traffic information database 12, the web information database 13, the broadcast database 14, the support information database 15, and the in-vehicle camera 21. , CAN signal acquisition unit 22, own vehicle position determination unit 31, road information acquisition unit 32, video analysis recognition unit 33, moving place determination unit 42, own vehicle state determination unit 44, presentation information selection unit 46, destination setting unit 51 In addition to the information presentation unit 52, a belonging / location information acquisition unit 101 that acquires an article, a movement purpose, and a movement destination that an occupant on the own vehicle has brought into the own vehicle, and general information Carrying items / moving eyes that are stored based on the article (carrying goods), movement purpose, and movement destination that the occupant of the own vehicle has brought into the vehicle. A database 102, a decoration state / movement location storage unit 103 in which an occupant of the own vehicle created based on personal information stores vehicle carry-on items, a movement purpose, and a movement destination as data, and current movement of the own vehicle And a movement purpose estimation unit 104 for estimating a purpose and a movement destination. That is, this movement purpose estimation apparatus includes a belonging / movement purpose database 102 instead of the decoration state / movement purpose database 16 in the movement purpose estimation apparatus shown as the first embodiment, and the decoration state / location information acquisition unit 41 includes Instead, a belonging / location information acquisition unit 101 is provided, a decoration state / movement location storage unit 103 is provided instead of the decoration state / movement location storage unit 43, and a decoration state / movement location is provided instead of the movement purpose estimation unit 45. A storage unit 104 is provided.

  The property / location information acquisition unit 101 creates the property / movement purpose database 102, the web information data read from the web information database 13, the broadcast data read from the broadcast database 14, and the support read from the support information database 15. On the basis of the information data and the like, an occupant riding in the host vehicle acquires a carry-on item, a movement purpose, and a movement destination. The personal belongings / location information acquisition unit 101 stores the personal belongings / location information indicating the acquired items carried by the occupant, the purpose of movement, and the destination of movement in the personal belongings / location database 102.

  The belonging / moving purpose database 102 includes, for example, a hard disk, and stores, as data, the moving purpose and the moving destination corresponding to the carry-on goods based on the belonging / location information acquired by the belonging / location information acquiring unit 101. To do. That is, the possession / movement purpose database 102 is based on general information such as web information data read from the web information database 13, broadcast data read from the broadcast database 14, and support information data read from the support information database 15. Created. The belonging / movement purpose database 102 may be mounted on the own vehicle or on a predetermined information distribution center side for distributing information. The data stored in the belonging / movement purpose database 102 is read out by the movement purpose estimation unit 104.

  The belonging / moving place storage unit 103 is composed of, for example, a hard disk and the like, based on the recognition information supplied from the video analysis recognition unit 33 and the moving place information supplied from the moving place determination unit 42. The corresponding movement purpose and movement destination are stored as data. That is, the personal belongings / moving place storage unit 103 is different from the personal belongings / moving purpose database 102, that is, personal information such as recognition information supplied from the video analysis recognition unit 33 and moving place information supplied from the moving place determination unit 42, that is, It is created based on information unique to the vehicle. The data stored in the belonging / movement location storage unit 103 is read by the movement purpose estimation unit 104.

  The movement purpose estimation unit 104 compares the recognition information supplied from the video analysis recognition unit 33 with the data stored in the property / movement object database 102 and the property / movement location storage unit 103, and based on the comparison result. Thus, the current travel destination and the travel destination of the host vehicle are estimated from the locations where the host vehicle has reached as the travel destination in the past. The movement purpose estimation unit 104 supplies the estimated movement purpose information to the presentation information selection unit 46 and the destination setting unit 51.

[Operation of moving purpose estimation device]
Such a movement purpose estimation device including each unit estimates a movement purpose according to a vehicle carry-in according to a series of procedures as shown in FIG.

  First, as shown in FIG. 7, the movement purpose estimation apparatus determines whether or not there is a new vehicle-carrying item brought into the host vehicle in step S61. That is, the movement purpose estimation device does not change the movement purpose and the movement destination until a new vehicle carry-on item is recognized. In addition, in order to determine the presence or absence of a new vehicle carry-in apparatus, the movement purpose estimation device acquires information on an IC tag that records an image of an occupant's home and information on an article at home, and is configured to be at home. There may be. In other words, the vehicle carry-in item includes an IC tag, and the movement purpose estimation device includes an IC tag reader. Then, the movement purpose estimation device performs processing so as not to change the movement purpose and the movement destination ground when the taken-out article is the same as the previous one. Actually, there are rare cases where the items brought in are exactly the same, but there are advantages such as saving the labor of the processing apparatus.

  Here, when the movement purpose estimation device recognizes that there is a new carry-in item brought into the host vehicle, in step S62, the in-vehicle image (video) captured by the in-vehicle camera 21 is acquired. In step S63, the image analysis recognition unit 33 extracts the shape information of the carry-on goods. The movement purpose estimation device stores the shape information recognized by the video analysis recognition unit 33 in the belongings / movement location storage unit 103 and is recognized by the video analysis recognition unit 33 by the movement purpose estimation unit 104 in step S64. Compared with the general shape information stored in the belonging / movement object database 102 and the shape information stored in the belonging / movement location storage unit 103, the movement information and the movement object having a high degree of similarity. After extracting the ground, the process proceeds to step S67.

  Further, in step S65, the movement purpose estimation apparatus extracts information such as the color and pattern of the article by the video analysis recognition unit 33. The movement purpose estimation device stores information such as colors and patterns recognized by the video analysis recognition unit 33 in the belongings / movement location storage unit 103, and in step S66, the movement purpose estimation unit 104 causes the video analysis recognition unit 33 to store the information. Information on colors and patterns recognized by the information, information on general colors and patterns stored in the possession / movement purpose database 102, and colors and patterns stored in the possession / movement location storage unit 43 After extracting a movement purpose and a movement destination having a high degree of similarity compared to the information such as the information, the process proceeds to step S67.

  Note that at the beginning of use, when the movement purpose and the movement destination are estimated based on the carry-on goods, since the amount of information in the belonging / movement location storage unit 103 is small, the movement purpose and the movement purpose database 102 are used. The moving destination will be estimated. However, as the movement of the own vehicle increases, the amount of information in the belonging / movement location storage unit 103 also increases, and the movement frequency and the estimated frequency of the movement destination using the belongings / movement location storage unit 103 gradually increase. Will increase.

  Similarly to the first embodiment, the personal belongings / movement purpose database 102 is created based on general information. For example, in a wedding hall, flowers, suits for changing clothes, etc. correspond to places as items brought in by the vehicle. If it is a tennis court, rackets and balls etc. are registered as carry-on items according to the place, and if it is a shopping center, shopping bags and personal carts correspond to places as carry-in items. Registered. Furthermore, when creating the personal belongings / movement purpose database 102, information on the ornament catalog on the Internet is acquired, information indicating the purpose of use of the clothes and the target audience, price, material, color, pattern, size, etc. Is acquired, and information such as the sales area, the number of sales, and the sales start date is also acquired. Furthermore, when creating the personal belongings / moving purpose database 102, the personal belonging information of each place can be obtained using video such as digital terrestrial broadcasting in the same manner as information extraction from the video of the webcam. Get catalog information for TV shopping.

  Then, the movement purpose estimation device checks whether or not the acquired information has already been registered in the belongings / movement purpose database 102. If there is no registration, the location (position), the vehicle carry-in item, The movement purpose and the like in the case of being in that place are registered in the belongings / movement purpose database 102 in association with each other. On the other hand, if there is a registration, the movement purpose estimation device compares only the information indicating the location and the items brought in by the vehicle. Shifts to acquiring other information without additionally registering the acquired information. In addition, if the compared information is greatly different and it is determined that the information is “difference”, the movement purpose estimation device further includes the date and time including the day of the week and the morning / afternoon, the season, the weather, and the sex. After assigning detailed conditions such as age group, the location, the article, the movement purpose, and the like are associated with each other and registered in the belongings / movement purpose database 102. Note that the moving purpose estimation device uses not only the belonging / movement purpose database 102 but also information indicating belongings stored in the belonging / movement location storage unit 103 which is a database dedicated to the own vehicle by the video analysis recognition unit 33. By comparing with the recognized belongings, the estimation accuracy of the movement purpose and the movement destination is improved.

  When the movement purpose estimation device compares the property / movement purpose database 102 and the property / movement location storage unit 103 with each of the possession / movement object database 102 and the possession / movement location storage unit 103 in each of steps S64 and S66, the estimated movement is performed in step S67 in FIG. The high degree of fitness for the purpose and the moving destination is compared by the moving purpose estimation unit 104. Then, the moving purpose and the moving destination having the highest fitness are obtained, and the moving purpose and the moving destination are determined as the final estimation result in step S68. At this time, the movement purpose estimation unit 104 has less customary movement purposes and destinations such as weddings, funerals, business trips, dates, concerts and other events such as ceremonial occasions, commuting, shopping, pick-up Travel destinations and destinations that have the highest fitness compared to all destinations and destinations, including both customary destinations and destinations, such as hospital visits, lessons, sports, vehicle maintenance, etc. Determine.

  In step S69, the movement purpose estimation apparatus determines whether the movement purpose and the movement destination determined in step S68 coincide with the current position by the movement purpose estimation unit 104. Here, when the movement destination and the movement destination do not coincide with the current location, the process proceeds to step S71 and step S73. On the other hand, if the movement purpose and the movement destination coincide with the current location, the next step related to the current movement purpose, such as going home, is performed by the movement purpose estimation unit 104 in step S70, as in the first embodiment. The item to be moved is estimated as the movement purpose.

  When the movement purpose and the movement destination are estimated in this way, in step S71, map data read from the map database 11, traffic information data read from the traffic information database 12, web information data read from the web information database 13, broadcast Based on the broadcast data read from the database 14, the support information data read from the support information database 15, the own vehicle state information supplied from the own vehicle state determination unit 44, and the like, the movement purpose estimated by the movement purpose estimation unit 104. And select the presentation information suitable for the moving destination. In step S72, the information presenting unit 52 presents the presentation information selected in step S71 to the occupant who is on the vehicle as voice information, graphic information, character information, and the like.

  In step S73, the destination setting unit 51 automatically sets a destination when performing route guidance for the host vehicle, or changes and sets the destination if an inquiry is obtained from the driver. Then, a series of processing is completed.

  The movement purpose estimation device can appropriately estimate the movement purpose according to the carry-on goods of the occupant on the host vehicle according to such a series of procedures.

  Specifically, the movement purpose estimation device, for example, if the vehicle carry-on item is a tennis racket or the like, based on the shape and size of the racket, the manufacturer name written on the cover, the model number of the body, etc. You can estimate whether you are going to practice or playing tennis. Similarly, the movement purpose estimation device can estimate shopping as a movement purpose when a shopping bag dedicated to shopping is recognized, and when the contents of the shopping bag are empty, go to the store for shopping from now on. If the contents are estimated and the contents of the shopping bag are contained, the return home can be estimated for the purpose of movement. In this case, if the purchased item is fresh food, the movement purpose estimation device immediately estimates the return home as the purpose of movement. Consider the possibility of going shopping. The movement purpose estimation device further increases the estimation accuracy (accuracy) by taking into account other factors such as the date and time, such as a movement for going to another shopping if it is a weekend shopping. Can be improved. In addition, when the bag is recognized, if the bag is a commuter bag or a business trip bag, the movement purpose estimation device estimates a commute to a company or a business trip to a business partner as a movement purpose. In the case of a casket for a ceremonial occasion, a hotel or a funeral hall serving as a wedding hall is estimated for the purpose of movement. Furthermore, in addition to the bag dedicated to shopping, the movement purpose estimation device was held by the occupant so that the in-car camera 21 could capture (view) newspaper insert advertisements on the day of movement, the day before, and the day before. Since the possibility of going to the flyer store increases by bringing it into the host vehicle in a state, shopping can be estimated as a moving purpose, and the flyer store can be estimated as a moving destination. Furthermore, when the camping equipment or articles used at sea are brought into the vehicle, the movement purpose estimation device estimates the purpose of movement to sea bathing or camping, and provides souvenirs such as confectionery and home waste. In the case of long vacations that are installed in the host vehicle, homecoming can be estimated as the purpose of travel.

  In addition, the movement purpose estimation device uses an IC tag information in which an image of an occupant's home or information on an article at home is recorded in order to improve the accuracy of recognizing an article brought into the host vehicle. It is desirable to confirm from the video of the in-vehicle camera 21 whether the item taken out from the vehicle has been brought into the host vehicle. Thereby, according to the movement purpose estimation device, it is possible to narrow down the objects to be recognized, so that the recognition accuracy can be improved.

  Furthermore, according to the movement purpose estimation device, in addition to the information on the items brought in by the vehicle, the estimation accuracy of the movement purpose and the movement destination can be further improved by taking into account the decoration state of the passenger as described in the first embodiment. It becomes possible to improve. For example, the movement purpose estimation device is a tennis racquet when the carry-on item is a tennis racket. In such a case, it is possible to accurately estimate playing tennis for the purpose of movement. At this time, the movement purpose estimation device can also estimate whether to go to tennis practice or to go to a tennis game based on the degree of occupant clothing. In addition, the movement purpose estimation device can narrow down the movement purpose to camping when the article brought into the host vehicle is camping equipment and the occupant's clothes are also for camping, If you have a large suitcase and wear a waist pouch, you can assume that you are traveling to the airport on an overseas trip. Furthermore, the movement purpose estimation device estimates that a visit to a hospitalized family member or going to nursery is a movement purpose when a lot of underwear, food for insertion, reading materials, games, etc. are brought into the vehicle. be able to. At this time, the movement purpose estimation apparatus can improve the estimation accuracy by considering the meeting time and the movement time.

  As described above, according to the movement destination estimation device, it is possible to estimate a movement destination and a movement destination that were not possible in the past, and as a result, even a movement destination with less habituality is estimated. be able to. Further, according to the movement purpose estimation device, even when the movement destination is the same, the movement purpose is effective even when the movement purpose is different, and it is possible to provide information suitable for the situation of the occupant.

[Effects of Second Embodiment]
As described above in detail, according to the movement purpose estimation device shown as the second embodiment of the present invention, the movement purpose and the movement purpose of the own vehicle based on the article recognized by imaging the goods brought in the vehicle. Since the ground is estimated, the movement purpose and the movement destination of the subject vehicle can be estimated with high accuracy, and the destination of behavior with less habit can be estimated with high accuracy.

  Further, according to this movement purpose estimation apparatus, an article taken out of the house is brought into the own vehicle using information on an IC tag in which an image of an occupant of the own vehicle or information on an article at home is recorded. Recognition by the video analysis and recognition unit 33 from the captured video, the processing accuracy for recognizing the article can be suppressed and the recognition accuracy can be improved. As a result, the estimation accuracy of the movement purpose and the movement destination can be improved. Can also be improved.

  The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and even if it is a form other than this embodiment, as long as it does not depart from the technical idea according to the present invention, the design and the like Of course, various modifications are possible.

It is a block diagram shown about the structure of the movement purpose estimation apparatus shown as 1st Embodiment of this invention. In the movement purpose estimation device shown as the first embodiment of the present invention, it is a flowchart showing a series of procedures when estimating the movement purpose according to the decoration state of the occupant riding the host vehicle. It is a flowchart shown about a series of procedures at the time of creating a decoration state / movement objective database in the movement objective estimation apparatus shown as 1st Embodiment of this invention. It is a figure shown about the structural example of a decoration state / movement objective database. In the movement purpose estimation apparatus shown as the first embodiment of the present invention, the decoration state recognized by the video analysis recognition unit and the decoration state stored in the decoration state / movement purpose database and the decoration state / movement location storage unit are shown. It is a flowchart shown about a series of procedures at the time of calculating similarity by comparing with information. It is a block diagram shown about the structure of the movement purpose estimation apparatus shown as 2nd Embodiment of this invention. In the movement purpose estimation device shown as the second embodiment of the present invention, it is a flowchart showing a series of procedures when an occupant on the own vehicle estimates a movement purpose according to a vehicle carry-on item.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Map database 12 Traffic information database 13 Web information database 14 Broadcast database 15 Support information database 16 Decoration state and movement purpose database 21 In-vehicle camera 22 CAN signal acquisition part 31 Own vehicle position determination part 32 Road information acquisition part 33 Image | video analysis recognition part 41 Decoration state / location information acquisition unit 42 Movement location determination unit 43 Decoration state / movement location storage unit 44 Vehicle state determination unit 45 Movement purpose estimation unit 46 Presentation information selection unit 51 Destination setting unit 52 Information presentation unit

Claims (7)

Imaging means for imaging the situation in the vehicle;
First storage means for storing the video imaged by the imaging means, the decoration state of the occupant of the vehicle recognized from the video image, and the movement purpose at that time in association with each other as a database;
A database that stores feature information including clothes and ornaments as a person's decoration state, and a movement purpose corresponding to the feature information, and is stored as a database created using information registered in a general-purpose network A second storage means;
A movement purpose estimating means for estimating a movement purpose of the vehicle,
The movement purpose estimation means includes
The video stored in the first storage means and the decoration state recognized from the video imaged by the imaging means are compared, and the first moving purpose of the vehicle is estimated based on the comparison result. , Estimating the second movement purpose of the vehicle based on the feature information stored in the second storage means,
When the first movement purpose and the second movement purpose are not significantly different from each other, the first movement purpose is preferentially estimated, and the first movement purpose and the second movement purpose are greatly different from each other. The movement purpose estimation apparatus characterized in that the decoration state recognized from the video imaged by the imaging means is recorded in the first storage means as a unique condition. Presentation information selection means for selecting presentation information corresponding to the movement purpose estimated by the movement purpose estimation means;
The presentation information selected by the presentation information selecting means, the moving object estimating apparatus according to claim 1, characterized in that it comprises an information presenting means for presenting the occupant of the vehicle. The movement purpose estimation apparatus according to claim 1 , further comprising destination setting means for setting a destination when performing route guidance of the vehicle based on the movement purpose estimated by the movement purpose estimation means. The destination setting means asks the occupant of the vehicle to set a destination when performing route guidance for the vehicle based on the movement purpose estimated by the movement purpose estimation means, and then moves the destination. The movement purpose estimation apparatus according to claim 3 , wherein the destination is set based on a purpose. The moving object estimating means, on the basis of the vehicle occupant carryons recognized from the captured image by the imaging means, of claims 1 to 4, characterized in that estimating the movement target of the vehicle The movement purpose estimation apparatus according to any one of the preceding claims. The movement purpose estimation means uses a pre-stored article image or IC tag information provided in the article among articles in the vehicle other than the vehicle structure recognized from the video imaged by the imaging means, 6. The moving purpose estimation apparatus according to claim 5 , wherein it is determined whether or not an article in the vehicle is a carry-on item. When the vehicle starts running, it captures the situation inside the vehicle,
Information stored in first storage means for storing the captured image , the decoration state of the vehicle occupant recognized from the image, and the purpose of movement at that time as a database, and human decoration A second database that stores feature information including clothes and ornaments as states and a movement purpose corresponding to the feature information, and that is stored as a database created using information registered in the general-purpose network. and have use of the information stored in the storage means, the first comparing the decorative state recognized from the image and the captured image stored in the storage means, the vehicle on the basis of the comparison result Estimating the first movement purpose of the vehicle and estimating the second movement purpose of the vehicle based on the feature information stored in the second storage means,
When the first movement purpose and the second movement purpose are not significantly different from each other, the first movement purpose is preferentially estimated, and the first movement purpose and the second movement purpose are greatly different from each other. The method for estimating the purpose of movement, wherein the decoration state recognized from the captured video is recorded in the first storage means as a unique condition.

Images