JP2016057801A - Charging processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system capable of performing charging to a vehicle at an arbitrary position by using a method for collating observation images which can be acquired from a vehicle during traveling with preliminarily stored reference images.SOLUTION: A charging processing system 1 for performing charging processing when a user vehicle 30 travels on a road includes: a reference image database for storing reference images preliminarily acquired in association with a plurality of charging positions set on the road; an observation image acquisition part 31 installed in the user vehicle 30 for acquiring observation images by photographing landscapes around the road on which the user vehicle 30 travels; a data processing part 33 for performing image matching between the stored reference images and the newly acquired observation images; and a charging processing part 35 for performing charging processing to the user vehicle 30 on the basis of the result of the image matching by the data processing part 33.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a system for charging a vehicle traveling on a road, and more particularly to a system capable of charging at an arbitrary position.

  As a system following the ERC (Electronic Loading Pricing) system of the DSRC (Dedicated Short Range Communication) system, a system that charges according to the travel position of the vehicle, that is, the travel distance, is promising. For this purpose, it is desired to charge with an accuracy of about ± 10 m and notify the driver of the fact of charging in real time. GPS (Global Positioning System) can be used to specify the traveling position of the vehicle.

  However, when this GPS positioning means is applied to an urban area where high-rise buildings that are the target areas of ERP are lined up, the GPS signal is blocked by the building or the overhead, or the GPS signal is irregularly reflected. To face. Therefore, in general, in an urban area where there are many high-rise structures, it is difficult to obtain the original positioning accuracy of GPS, and there is a possibility that the charging position may be shifted or charged with an incorrect amount. In addition, a decrease in GPS positioning accuracy occurs in an unspecified situation depending on the satellite arrangement, atmospheric conditions, and the like, and the magnitude of the error also varies with time. Therefore, it is difficult to ensure the accuracy of the billing position and the reliability of billing only with GPS.

In order to cope with a decrease in the accuracy of GPS in a high-rise building area, for example, it is assumed that compensation is performed by a gyroscope or an acceleration sensor, or that the position is corrected on the road using map information. For example, Patent Document 1 proposes that the position of the vehicle is specified by correcting it using information related to a positioning error that occurs due to the environment. Even if the GPS accuracy is reduced, the accuracy is maintained by the gyroscope and acceleration sensor for a certain period of time, but this correction involves integrating the values of the gyroscope and acceleration sensor, so the error increases over time. End up.
In addition, it is possible to estimate the self-position by inertial navigation even when the GPS is interrupted by using a positioning device such as a high-performance vehicle speed pulse or a high-performance gyroscope. However, these devices are too expensive to install in private user vehicles.

  Therefore, it is desirable for a system that charges according to the traveling position of the vehicle to be able to specify the position of the traveling vehicle without relying on GPS and without attaching expensive equipment. For example, in Patent Document 2, a surrounding image (observation image) at the current position of a moving object is collated with an image (reference image) prepared in advance in relation to a predetermined point by image matching. Based on this, it is proposed to calculate the current position of the moving object.

JP 2014-92986 A JP 2013-61270 A

However, Patent Literature 2 does not suggest a system for charging the vehicle, although the current position of the vehicle can be specified.
Therefore, an object of the present invention is to provide a system that can charge a vehicle at an arbitrary position by using a method of collating an observation image that can be acquired from a traveling vehicle with a reference image stored in advance.

  For this purpose, the billing processing system of the present invention is a billing processing system that performs billing processing when a user vehicle travels on a road, and is acquired in advance in association with a plurality of billing positions set on the road. A reference image storage unit that stores the reference image, an observation image acquisition unit that is provided in the user vehicle and that captures a landscape around a specific road on which the user vehicle runs and acquires an observation image, and a stored reference The image processing apparatus includes a data processing unit that performs image matching between the image and a newly acquired observation image, and a charging processing unit that performs charging processing on the user vehicle based on a result of image matching in the data processing unit.

  In the billing processing system of the present invention, the data processing unit can perform image matching with the observation image newly acquired for the selected reference image.

  In the charging processing system of the present invention, the data processing unit obtains the similarity between the reference image and the observed image by image matching, and the charging processing unit performs charging processing at a charging position corresponding to the reference image based on the similarity. be able to.

  In the billing processing system of the present invention, the data processing unit estimates a relative positional relationship between a position where the reference image is photographed and a position where the observation image is photographed, for the reference image and the observation image to be image-matched. Then, the charging processing unit can perform the charging process at the charging position corresponding to the reference image based on the estimation result.

  In the charging processing system of the present invention, the data processing unit obtains a predicted passing time when the user vehicle passes the charging position corresponding to the reference image based on the reference image and the observed image, and the charging processing unit The user vehicle can be charged.

In the present invention, there are at least two methods for obtaining the predicted passage time.
First, the data processing unit is based on the distance to the charging position calculated based on the relative positional relationship between the position where the reference image corresponding to the charging position is captured and the position where the observation image is captured, Obtain the estimated passage time.
Second, the data processing unit calculates the estimated passage time based on the relative positional relationship between the position where the reference image corresponding to the charging position is captured and the position where the observation image is captured, and the traveling speed of the user vehicle. Ask.

  Some components in the accounting processing system of the present invention described above are realized as an in-vehicle device mounted on a user vehicle. That is, the present invention is an in-vehicle device mounted on a user vehicle that performs charging processing when the user vehicle travels on a road, and is a reference acquired in advance related to a plurality of charging positions set on the road. A data processing unit that performs image matching between an image and an observation image obtained by capturing a landscape around a road on which the user vehicle travels, and a charging process that charges the user vehicle based on a result of image matching in the data processing unit And a section.

  The vehicle-mounted device of the present invention can apply a more specific form in the accounting processing system of the present invention described above.

According to the billing processing system of the present invention, the billing position is set corresponding to the reference image information. Therefore, the billing position is specified by performing the matching process on the observed image information including the position information in the reference image information. Can do.
Therefore, the billing processing system of the present invention enables billing based on the position through which the user vehicle passes while traveling. In addition, since the billing processing system can identify the billing position without depending on the GPS, the billing processing can be performed without being affected by the deterioration of the accuracy of GPS. Also, even in areas where GPS is not available, such as inside tunnels and underground roads, for example, characters and other information are drawn on the wall of the tunnel, so that the position and other positions can be distinguished on the captured image. If possible, billing processing can be performed.

  In addition, the accuracy of the billing position specified in the billing processing system according to the present invention is determined by the imaging conditions of the image information of both the reference image information and the observation image information, and the matching method of the reference image information and the observation image information. Is known. On the other hand, the accuracy of GPS varies greatly depending on the position where the GPS signal is acquired. Therefore, according to the present invention, charging reliability can be ensured as compared to using GPS.

It is a figure which shows the concept of the accounting processing system which concerns on 1st Embodiment of this invention. In 1st Embodiment, it is a figure which shows a mode that the imaging | photography vehicle which acquires a reference image drive | works. In 1st Embodiment, it is a figure which shows a mode that the user vehicle which acquires an observation image drive | works. It is a figure which shows the outline | summary of a structure of the accounting processing system in 1st Embodiment. It is a figure which shows the content of the database which the server in 1st Embodiment hold | maintains. It is a figure which shows the mode of the matching process in the user vehicle in 1st Embodiment. It is a figure which shows the procedure of the process of the user vehicle in 1st Embodiment. It is a figure which shows the procedure of the process of the user vehicle in 2nd Embodiment. In 2nd Embodiment, it is a figure which shows the relationship between the position which acquired the observation image, and the position which acquired the reference image. It is a figure which shows the content of the database which the server in 2nd Embodiment hold | maintains. It is a figure explaining the ego motion (Ego-motion) estimation applied in 2nd Embodiment. In 2nd Embodiment, it is a figure explaining the method of improving the precision of position estimation. It is a figure which shows the content of the process which estimates the time which passes a charge position in 3rd Embodiment (1st method). It is a figure which shows the procedure of the accounting process in the user vehicle in 3rd Embodiment (1st method). It is a figure which shows the content of the process which estimates the time which passes a charge position in 3rd Embodiment (2nd method). It is a figure which shows the procedure of the accounting process in the user vehicle in 3rd Embodiment (2nd method).

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.
[First Embodiment]
The vehicle charging processing system 1 shown in FIGS. 1 to 3 is a system that performs charging processing on the vehicle according to the travel position of the user vehicle 30 in the charging target area Z without using GPS. In addition, the system 1 is a system that can perform billing processing according to the travel position of the user vehicle 30 without providing a large-scale structure like a conventional gantry.

  The accounting processing system 1 uses the observation image generated by the user vehicle 30 continuously shooting while traveling in the charging target area Z based on the reference image information generated by shooting the vehicle 10 traveling in the charging target area Z in advance. The traveling position of the user vehicle 30 is specified by performing a matching process on the image information. The billing processing system 1 performs billing processing on the user vehicle 30 when the identified travel position corresponds to a position where billing is to be performed. As shown in FIG. 1, the charging processing system 1 includes a host 20 in addition to the photographing vehicle 10 and the user vehicle 30, and the host 20 holds reference image information.

  In constructing the billing processing system 1, the photographing vehicle 10 obtains a reference image by photographing a surrounding landscape while traveling on the road R in the billing target area Z as shown in FIG. For this purpose, the photographing vehicle 10 includes a reference image acquisition unit 11, a positioning unit 13, a reference image storage unit 15, and a transmission / reception unit 17, as shown in FIG.

The reference image acquisition unit 11 includes a camera that acquires a reference image by photographing all directions around the photographing vehicle 10 at equal intervals, for example, 10 m or 5 m. The photographing vehicle 10 measures the latitude and longitude of the photographing vehicle 10 by the positioning unit 13 while acquiring the reference image. The captured reference image is stored in the reference image storage unit 15 after being associated with the latitude and longitude (position information) at which the image was captured. The stored information (hereinafter referred to as reference image information) is sent to the host 20 via the transmission / reception unit 17. The reference image information acquired for all roads R that are to be charged using the single or plural photographing vehicles 10 is sent to the host 20.
The photographing vehicle 10 accurately measures latitude and longitude by the positioning unit 13 including a high-performance positioning device.

As illustrated in FIG. 4, the host 20 includes a storage unit 21, a data processing unit 25, and a transmission / reception unit 27.
The host 20 acquires reference image information sent from the photographing vehicle 10 via the transmission / reception unit 27. This reference image information is stored in the storage unit 21 after being converted into a feature value by the data processing unit 25 for use in matching with the observed image. As shown in FIG. 5, for example, the storage unit 21 includes a reference image information database 22, a charging information database 23, and a customer information database 24.
As shown in FIG. 5, the reference image information database 22 includes image data (images # 1, # 2, # 3...) As feature amounts and position data ((Lat1, Lon1), ( Lat1, Lon2), (Lat1, Lon3) ...) are stored in table format information.
Further, as shown in FIG. 5, the billing information database 23 is a table in which billing location data indicating the billing location in latitude and longitude and a billing amount that is the amount billed in the billing location data are associated with each other. Stores format information. The billing position (data) and the billing amount (data) stored in the billing information database 23 are set when the billing processing system 1 is operated, and billing information in the storage unit 21 is transmitted from an appropriate input unit via the transmission / reception unit 27. Input to the database 23.
Further, the customer information database 24 stores the identification information (ID) assigned to each user vehicle 30 mounted on the user vehicle and the charging status in association with each other. Each time a new charge is made, the stored contents of the customer information database 24 are updated.

  The data processing unit 25 converts the reference image information sent from the photographing vehicle 10 into a feature amount, and then stores it in the reference image information database 22 of the storage unit 21. The data processing unit 25 reads information stored in the reference image information database 22, the billing information database 23, and the customer information database 24 in the storage unit 21, and transmits the information to the user vehicle 30 via the transmission / reception unit 27. .

Next, as shown in FIG. 3, the user vehicle 30 performs a matching process between information (observation image information) related to the surrounding landscape captured while traveling in the chargeable area Z and reference image information stored in the host 20. Thus, if it is recognized that the user vehicle 30 passes the charging position, the charging process is performed. Therefore, as shown in FIG. 4, the user vehicle 30 includes an observation image acquisition unit 31, a data processing unit 33, a charging processing unit 35, and a transmission / reception unit 37. In addition, although the component part which consists of the observation image acquisition part 31, the data processing part 33, the charge process part 35, and the transmission / reception part 37 can be mounted in the user vehicle 30 as what is called an onboard equipment unit, here, it is only the user vehicle 30 and write.
The observation image acquisition unit 31 captures the surrounding scenery while the user vehicle 30 is traveling in the chargeable area Z, and sends the captured image information to the data processing unit 33. Note that the observation image converted into the feature amount is referred to as observation image information.

The data processing unit 33 performs a matching process between the observation image information captured by the observation image acquisition unit 31 and the reference image information stored in the storage unit 21 of the host 20, and specifies the passage of the charging location as a result of the matching process. To do. More specifically, it is as follows.
The data processing unit 33 takes in the observation image captured by the observation image acquisition unit 31 and converts it into a feature amount. On the other hand, the data processing unit 33 takes in all the reference image information # 1 to #N stored from the reference image information database 22 of the host 20, matches the latest observed image information just taken, and The similarity between the information is calculated, and the most similar reference image information is specified. For example, if it is determined that the reference image information # 3 in the reference image information database 22 has the highest similarity, the data processing unit 33 is set to the position (Lat1, Lon3) where the reference image information # 3 is acquired. Specify to approximate. If the reference image information whose position is approximated is specified, the data processing unit 33 next refers to the billing information database 23 and confirms the presence / absence of data that matches the position of the reference image information. Here, since the position (Lat1, Lon3) corresponding to the reference image information # 3 is stored in the charging information database 23, the data processing unit 33 recognizes that the reference image information # 3 is the charging position, and The corresponding charge amount is acquired from the charge information database 23.

  The classification of the reference image information database 22, the billing information database 23, and the customer information database 24 is merely an example. For example, the reference image information database 22 and the billing information database 23 may be integrated. In short, it suffices if the billing location and the billing amount at the billing location can be specified.

The billing processing unit 35 executes billing processing when the data processing unit 33 specifies the billing location.
As the billing process, the identified billing location and the billing amount corresponding to this billing location are transmitted to the host 20 in association with its own ID. The host 20 stores the acquired information related to the billing process in the corresponding ID recording area of the customer information database 24.
Further, as a billing process, the fact of billing is displayed on a display (not shown) of the user vehicle 30. As a display method, only the charge amount can be displayed, or the charge position and the charge amount can be written together. In addition to displaying the accounting fact as character information on the display, the accounting fact can be notified by voice.

[Operation of Billing Processing System 1]
Now, the operation of the billing processing system 1 having the above configuration will be described with reference to FIG. It is assumed that predetermined reference image information, billing information, and customer information are stored in the reference image information database 22, billing information database 23, and customer information database 24 in the storage unit 21 of the host 20, respectively.

Now, it is assumed that the user vehicle 30 including the vehicle-mounted device unit is traveling in the charge target area Z.
While traveling, the user vehicle 30 captures an image of the surrounding landscape by the observation image acquisition unit 31 of the user vehicle 30 and converts it into a feature amount to acquire observation image information (S101 in FIG. 7).
Once the user vehicle 30 has acquired the observation image information, the data processing unit 33 performs matching processing (S103 in FIG. 7). The specific contents of the matching process are as described above, but the reference image information having the highest similarity is specified by this matching process. For example, in FIG. 6, if the reference image information # 3 has the largest similarity to the acquired observation image information, the data processing unit 33 specifies the reference image information # 3 as the current position of the user vehicle 30.

Next, the user vehicle 30 determines whether or not the specified position of the user vehicle 30 corresponds to the charging position (S105 in FIG. 7).
This determination is made by collating the specified position of the user vehicle 30 with the billing information read from the billing information database 23 of the host 20 by the data processing unit 33. If the corresponding billing location is verified (S105 Y in FIG. 7), the data processing unit 33 next performs billing processing (S107 in FIG. 7). The specific contents of the billing process are as described above. When the accounting process is completed, new observation image information is acquired, and thereafter, the procedure after the matching process is repeated.
On the other hand, if the corresponding billing position is not verified (S105 N in FIG. 7), the observation image information is newly acquired, the new observation image information is also acquired, and thereafter, the procedure after the matching process is repeated.

[Effect of billing system 1]
As described above, since the charging processing system 1 sets the charging position corresponding to the reference image information, the charging position is specified by matching the observation image information including the position information with the reference image information. be able to.
Accordingly, the billing processing system 1 enables billing based on the position through which the user vehicle 30 passes while traveling. In addition, since the accounting processing system 1 can specify the accounting location without depending on the GPS, the accounting processing can be performed without being affected by the deterioration of the accuracy of GPS. Also, even in areas where GPS is not available, such as inside tunnels and underground roads, for example, characters and other information are drawn on the wall of the tunnel, so that the position and other positions can be distinguished on the captured image. If possible, billing processing can be performed.
In addition, the accuracy of the billing position specified in the billing processing system 1 is determined by the imaging conditions of the image information of both the reference image information and the observation image information, and the matching method of the reference image information and the observation image information. It can be said. On the other hand, the accuracy of GPS varies greatly depending on the position where the GPS signal is acquired. Therefore, according to the billing processing system 1, billing reliability can be ensured as compared to using the GPS.

[Example of change]
The preferred embodiment of the present invention has been described above by taking the charging processing system 1 as an example. However, the configuration of the charging processing system 1 can be changed without departing from the gist of the present invention.
For example, the billing processing system 1 stores the reference image location information and billing information in the storage unit 21 of the host 20 that exists in a remote place as a separate body from the user vehicle 30, but the capacity of the storage medium is allowed. For example, the reference image place information and the billing information can be stored in the user vehicle 30.
Further, the billing processing system 1 performs matching processing and billing processing in the user vehicle 30. However, according to the present invention, an observation image that is an image around the user vehicle 30 captured by the user vehicle 30 is transmitted to the host 20, and the host 20 performs characterization of the observation image, matching processing, and billing processing. it can. That is, the present invention only needs to be able to perform matching processing and billing processing as a whole of the billing processing system 1 including the user vehicle 30.

  The accounting processing system 1 also stores positions that do not correspond to the accounting position in the reference image information database 22 at present, but this makes it possible to respond quickly to changing the accounting position later. For example, # 3 is currently set as the charging position, but when the charging position is changed from # 3 to # 4 later, there is no need to take a reference image corresponding to # 4 again. However, according to the present invention, only the observation image information corresponding to the charging position can be stored in the reference image information database 22. In this case, the charging information can be included in the observation image information, and it is not necessary to provide the charging information database 23 individually.

  In addition, the accounting processing system 1 targets all reference image information to be subjected to matching processing with observation image information, but performs matching processing only on a part of the reference images stored in the reference image information database 22. Can do. For example, it is possible to extract several billing positions that the user vehicle 30 can pass next based on the billing positions specified in advance, and to perform matching processing by focusing on the extracted billing positions. For example, if # 3 is specified in advance as the billing position, and if the billing position that can be passed next is limited to # 5, # 11, and # 14 due to road connection, # 5, # 11 The observation image information can be matched only to the reference image information corresponding to 11 and # 14. Further, when the user vehicle 30 includes a car navigation system based on GPS, only reference image information having a predetermined distance relationship with the self position of the user vehicle 30 obtained from the GPS signal is extracted from the reference image information database 22. Thus, it can be a target of matching processing.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIGS.
In the first embodiment, the accuracy of the position specified by the matching process depends on the interval at which the reference image is captured. That is, when the reference images are photographed at intervals of 10 m as in the first embodiment, the obtained accuracy is limited to a range of ± 5 m at the maximum. Therefore, in the second embodiment, a method for specifying the position with higher accuracy than the interval between the reference images is proposed.

  The charging processing system 2 according to the second embodiment has the basic configuration of including the photographing vehicle 10, the host 20, and the user vehicle 30 on which the vehicle-mounted unit is mounted. Common to system 1. Therefore, the charging processing system 2 will be described below with a focus on differences from the charging processing system 1.

First, an outline of the accounting processing system 2 will be described with reference to FIGS.
As shown in FIG. 8, the billing processing system 2 determines the relative positional relationship between the position where the observation image information is photographed and the position where the reference image information is photographed in the matching process (corresponding to S103 in FIG. 7). Calculate (S203 in FIG. 8). This calculation is performed by the data processing unit 33 (FIG. 4) as follows. That is, the data processing unit 33 takes in all the reference image information # 1 to #N stored from the reference image information database 22 of the host 20, matches the latest observed image information that has just been photographed, and obtains image information. Distances ΔL1 to N, which are relative positional relationships between them, are calculated. Then, the data processing unit 33 specifies reference image information # 1 to #N having the smallest calculated distances ΔL1 to ΔN. For example, if it is determined that the distance ΔL3 to the reference image information # 3 in the reference image information database 22 is the shortest, the data processing unit 33 determines that its position is the position (Lat1, Lon3) from which the reference image information # 3 is acquired. ) And the distance ΔL3 is specified. If the reference image information whose position is approximated is specified, the data processing unit 33 next refers to the billing information database 23 and confirms the presence / absence of data that matches the position of the reference image information. Here, since the position (Lat1, Lon3) corresponding to the reference image information # 3 is stored in the charging information database 23, the data processing unit 33 recognizes that the reference image information # 3 is the charging position, and The corresponding charge amount is acquired from the charge information database 23.

For the calculation of the relative positional relationship described above, ego-motion estimation, which is processing for estimating the three-dimensional movement of the camera in the environment using an image, may be applied. However, in order to perform the above processing, it is necessary to provide scale information as reference image information, as shown in FIG. In FIG. 10, the accounting information database 23 and the customer information database 24 are not shown.
As shown in FIG. 11, the scale information is a distance (actual dimension: S _A-B ) of a line segment connecting any two points A and B to the reference image information. Since the relative positional relationship calculation method based on egomotion estimation is known as disclosed in Non-Patent Document 1, for example, the details are omitted, but it is observation image information to be subjected to matching processing, and two arbitrary points A , two points a corresponding to B, the line segment connecting the b the actual dimension S _a-b by regarding the S _a-B, it calculates a relative positional relationship. In FIG. 11, P indicates the shooting position of the reference image, and D indicates the shooting position of the observation image.

[Effects of Second Embodiment]
Since it is as above, according to 2nd Embodiment, there can exist the following effects.
First, in the second embodiment, the relative positional relationship between the position where the observed image information is captured and the position where the reference image information is captured is calculated. Therefore, the second embodiment is estimated with higher accuracy than the interval at which the reference image is captured. The billing location can be identified based on the location.
For example, when there are a plurality of lanes in which the user vehicle 30 travels, the relative positional relationship is calculated even when the lane in which the reference image is captured and the lane in which the user vehicle 30 travels are different. Thus, the billing location can be specified with high accuracy.

  Further, according to the second embodiment, even if the number of reference image information is reduced, the distance to the billing position can be estimated, so the image is taken with the photographing vehicle 10 and stored in the reference image information database 22 of the host 20. The number of reference image information can be reduced.

[Modification Example of Second Embodiment]
The accuracy of position estimation according to the second embodiment described above can be further improved. Hereinafter, this method will be described.
This accuracy improvement method calculates a relative positional relationship between a position where a plurality of omnidirectional images are captured and an in-vehicle camera in an image matching process, and obtains a plurality of pieces of information as relative positions at each time. By integrating, typically averaging, a plurality of obtained information, the influence of errors generated in each image matching process is reduced. Hereinafter, a more specific description will be given with reference to FIG.

FIG. 12 shows the positions at which the reference images are taken as P1, P2, P3, and P4 on the road R. The user vehicle 30 travels on the road R to take an observation image, and performs matching processing. And Here, an example of processing the observed image $ taken at the position D1 is shown.
The relative positional relationship between the observed image captured at the position D1 and the reference image captured at each of the positions P1, P2, P3, and P4 is calculated. In addition, each space | interval of position P1, P2, P3, P4 shall be (alpha) (for example, 10 m) correctly.

For example, as shown in FIG. 12, the calculation result is a distance ΔL1 between the position D1 and the position P1, a distance ΔL2 between the position D1 and the position P2, a distance ΔL3 between the position D1 and the position P3, and a distance ΔL3 between the position D1 and the position P4. It shall be calculated | required as L4. As described above, the distance between the position D1 and the position P1 can be estimated only by ΔL1, but here, the distance Δ between the position D1 and the position P1 is also considered in consideration of the calculation results of ΔL2, ΔL3, and ΔL4. L is calculated. For this purpose, as shown in the following formula (1), ΔL1 is averaged by adding a value obtained by subtracting an accurate distance (α, 2α, 3α) from position P1 to each of positions P2, P3, P4.
ΔL = (ΔL1 + (ΔL2-α) + (ΔL3-2α) + (ΔL4-3α)) / 4 Formula (1)

By averaging as described above, the influence of errors generated in the matching process can be reduced, so that the position estimation accuracy in the second embodiment can be further improved.
In the present invention, weighting can be applied in addition to simple arithmetic averaging. The weighting is, for example, that the coefficient is decreased as the distance increases, and the coefficients multiplied by the terms ΔL1, ΔL2, ΔL3, and ΔL4 are sequentially decreased.

[Third Embodiment]
Next, a third embodiment of the present invention will be described with reference to FIGS.
In the first embodiment and the second embodiment, it is possible to determine that the image is most similar to “the scenery of the charging position” only after passing through the charging position. As a result, at the moment when the user vehicle 30 passes the billing position, the user vehicle 30 cannot determine the passage and perform billing processing.
The third embodiment provides a method capable of performing charging processing at the moment when the charging position is passed. The third embodiment proposes two methods, a first method and a second method.
The first method calculates the distance to the charging position of the user vehicle 30 in a time series based on the reference image information and the observation image information used in the second embodiment, so that the user vehicle 30 determines the charging position. Predict the time to pass. The second method predicts the time when the user vehicle 30 passes through the charging position by calculating the traveling speed of the user vehicle 30 from the observed image captured by the user vehicle 30. Hereinafter, the first method and the second method will be described in this order.

[First method]
In the first method, the relative positional relationship of the user vehicle 30 is calculated by performing a matching process on the reference image information in which the charging position is specified and the observed image information. The calculation of the relative positional relationship is performed in the same manner as in the second embodiment. When this relative positional relationship is calculated, the distance to the charging position P can be calculated at a point before the charging position P. When this process is performed in time series, the time series change of the distance to the charging position P can be calculated. However, when this time series change is extrapolated, the user vehicle 30 is connected to the charging position P before reaching the charging position P. The time to pass can be predicted. Therefore, by using this predicted passage time, the user vehicle 30 can perform billing processing at the moment when it passes the billing position P. Hereinafter, the first method will be described more specifically with reference to FIGS. 13 and 14.

  The billing processing system 3-1 according to the first method is the same as that of the first embodiment in that it includes a basic configuration that includes a photographing vehicle 10, a host 20, and a user vehicle 30 on which an onboard unit is mounted. Common with the accounting processing system 1. Therefore, in the following, the charging processing system 3-1 will be described focusing on the differences from the charging processing system 1, that is, the processes related to S205 and S206 in the procedure shown in FIG.

As shown in FIG. 13A, it is assumed that the user vehicle 30 travels on the road toward the charging position P, and the observation image acquisition unit 31 of the user vehicle 30 continuously shoots the surrounding scenery. . Then, suppose that the matching processing between the observation image information ($ 1 to $ 4) based on the scenery photographed at the observation positions D1, D2, D3, and D4 on the road R and the observation image information (#N) based on the charging position P is performed. It shall be performed in order. This matching process is performed by the data processing unit 33.
By this matching process, the relative positional relationship between the observation position D1 and the charging position P is calculated, and then the distance ΔL1 between the observation position D1 and the charging position P is obtained. FIG. 13B shows this on a graph in which the horizontal axis represents time and the vertical axis represents distance.
Similarly, a distance ΔL2 between the observation position D2 and the charging position P, a distance ΔL3 between the observation position D3 and the charging position P, and a distance ΔL4 between the observation position D4 and the charging position P are time-sequentially. Ask. Then, a time series change of the distance to the charging position P is obtained, and when these are extrapolated in order, the charging position P is set as shown by the white arrows in FIGS. 13 (c), (d), and (e). The passing time can be predicted in advance.

  As shown in FIGS. 13B to 13E, as the user vehicle 30 travels, the advance prediction of the time of passing the charging position P is updated each time the observation positions D1, D2, D3, D4 are reached in order. To do. However, the present invention does not necessarily require the updating of the prediction, and the prediction of the time that passes the charging position P only at a specific observation position with respect to the charging position P, for example, only at the observation position D4 close to the charging position P. Can also be done. Of course, in the region close to the charging position P, the fluctuation of the traveling speed of the user vehicle 30 may be large. Therefore, a time series change is observed over a certain period, and the charging position P is passed in consideration of this change. It is preferable to predict the time.

[Second method]
The second method is common to the first method in that the distance to the charging position P of the user vehicle 30 is calculated and the user vehicle 30 predicts the passing time of the charging position P. However, the second method uses the traveling speed of the user vehicle for this prediction.

  The billing processing system 3-2 according to the second method is a billing processing system 3-3 in that it has a basic configuration including a photographing vehicle 10, a host 20, and a user vehicle 30 on which an onboard unit is mounted. 1 and common. Therefore, in the following, the charging processing system 3-2 will be described focusing on the differences from the charging processing system 3-1, that is, the processes related to S202, S204, and S205 in the procedure shown in FIG.

As shown in FIG. 15A, it is assumed that the user vehicle 30 travels on the road toward the charging position P, and the observation image acquisition unit 31 of the user vehicle 30 continuously shoots the surrounding scenery. . Then, suppose that the matching processing between the observation image information ($ 1 to $ 4) based on the scenery photographed at the observation positions D1, D2, D3, and D4 on the road and the observation image information (#N) based on the charging position P is sequentially performed. Assumed to be performed. This matching process is performed by the data processing unit 33.
By this matching process, the distance ΔL1 between the observation position D1 and the charge position P, the distance ΔL2 between the observation position D2 and the charge position P, the observation position D3 and the charge position P, as in the first method. A distance ΔL3 between them, and a distance ΔL4 between the observation position D4 and the charging position P are obtained in time series.

In parallel with the above calculation of the distance, the data processing unit 33 calculates the traveling speed v of the user vehicle 30 from the observed images taken continuously. This traveling speed v can be calculated by, for example, visual odometry. In order to improve the accuracy of the traveling speed v calculated by visual odometry, a reference image acquisition unit 11 can be provided to capture a stereo image. In addition to obtaining from the observation image, the traveling speed v can also be a speed obtained by a device provided in the user vehicle 30, for example, an acceleration sensor or an encoder.
The data processing unit 33 obtains the traveling speed v1 at the observation position D1, the traveling speed v2 at the observation position D2, the traveling speed v3 at the observation position D3, and the traveling speed v4 at the observation position D4 in time series.

The data processing unit 33 can predict the time of passing the charging position P in advance by fusing the calculated distance ΔL1 and the traveling speed v1 using a stochastic model such as a Kalman filter. it can.
Similarly, at the observation position D2, the observation position D3, and the observation position D4, as shown in FIGS. 15 (c), (d), and (e), it is possible to predict in advance the time of passing the charging position P. .

  As shown in FIGS. 15 (b) to 15 (e), every time the observation positions D1, D2, D3, and D4 are reached in order as the user vehicle 30 travels, the advance prediction of the time passing through the charging position P is updated. To do. However, the present invention does not necessarily require the updating of the prediction, and the prediction of the time that passes the charging position P only at a specific observation position with respect to the charging position P, for example, only at the observation position D4 close to the charging position P. Can also be done. Of course, in the region close to the charging position P, the fluctuation of the traveling speed of the user vehicle 30 may be large. Therefore, a time series change is observed over a certain period, and the charging position P is passed in consideration of this change. It is preferable to predict the time.

  The billing processing system 3-2 uses the traveling speed v of the user vehicle 30 in predicting the passage time. The traveling speed v can vary greatly if the traveling distance is long, but if the traveling speed v is a relatively short distance including the charging position P, it can be considered that the variation is small, so that the passage time can be predicted. As the traveling speed v, an average traveling speed for a predetermined period can be used. For example, the traveling speed v1 applied at the observation position D1 is the average value of the traveling speed v in a predetermined period until reaching the observation position D1, and the traveling speed v2 applied at the observation position D2 is between the observation position D1 and the observation position D2. It is possible to set the average value of the traveling speed of Note that the traveling speed v of the user vehicle 30 is usually measured and recorded by the car navigation system, so that data may be used.

The preferred embodiments of the present invention have been described above. However, the configuration described in the above embodiments can be selected or changed to other configurations as appropriate without departing from the gist of the present invention. .
For example, the host 20 can have the functions of the data processing unit 33 and the charging processing unit 35 provided in the user vehicle 30. Conversely, the user vehicle 30 can be provided with the function of the data processing unit 25 of the host 20.

DESCRIPTION OF SYMBOLS 1 Charge processing system 10 Shooting vehicle 11 Reference image acquisition part 13 Positioning part 15 Reference image storage part 17 Transmission / reception part 20 Server 21 Storage part 22 Reference image information database 23 Charge information database 24 Customer information database 25 Data processing part 27 Transmission / reception part 30 User Vehicle 31 Observation image acquisition unit 33 Data processing unit 35 Charge processing unit 37 Transmission / reception unit R Road Z Chargeable area

Claims (14)

A charging processing system that performs charging processing when a user vehicle travels on a road,
A reference image storage unit that stores reference images acquired in advance in association with a plurality of charging positions set on the road;
An observation image acquisition unit that is provided in the user vehicle and acquires an observation image by photographing a landscape around the road on which the user vehicle runs;
A data processing unit that performs image matching between the stored reference image and the newly acquired observation image;
A charging processing unit for charging the user vehicle based on the result of the image matching in the data processing unit;
A billing processing system comprising: The data processing unit
The image matching with the observation image newly acquired for the selected reference image is performed.
The accounting processing system according to claim 1. The data processing unit
Obtaining the similarity between the reference image and the observed image by the image matching;
The billing processing unit
Charging processing is performed at the charging position corresponding to the reference image based on the similarity;
The billing processing system according to claim 1 or 2. The data processing unit
With respect to the reference image and the observation image to be image-matched, a relative positional relationship between a position where the reference image is captured and a position where the observation image is captured is estimated,
The billing processing unit
Based on the estimation result, charging processing is performed at the charging position corresponding to the reference image.
The billing processing system according to claim 1 or 2. The data processing unit
Based on the reference image and the observation image, obtain a predicted passage time when the user vehicle passes through the charging position corresponding to the reference image,
The billing processing unit
The user vehicle is charged at the estimated passage time.
The billing processing system according to any one of claims 1 to 4. The data processing unit
The estimated passage time is obtained based on a distance to the charging position calculated based on a relative positional relationship between the position where the reference image corresponding to the charging position is captured and the position where the observation image is captured. ,
The billing processing system according to claim 5. The data processing unit
Obtaining the predicted passage time based on a relative positional relationship between a position where the reference image corresponding to the charging position is photographed and a position where the observation image is photographed, and a traveling speed of the user vehicle;
The billing processing system according to claim 5 or 6. An on-vehicle device mounted on the user vehicle that performs billing processing when the user vehicle travels on a road,
A data processing unit that performs image matching between a reference image acquired in advance in association with a plurality of billing positions set on the road and an observation image obtained by photographing a landscape around the road on which the user vehicle travels; ,
A charging processing unit for charging the user vehicle based on the result of the image matching in the data processing unit;
A vehicle-mounted device comprising: The data processing unit
The image matching with the observation image newly acquired for the selected reference image is performed.
The vehicle-mounted device according to claim 8. The data processing unit
Obtaining the similarity between the reference image and the observed image by the image matching;
The billing processing unit
Charging processing is performed at the charging position corresponding to the reference image based on the similarity;
The vehicle-mounted device according to claim 8 or 9. The data processing unit
With respect to the reference image and the observation image to be image-matched, a relative positional relationship between a position where the reference image is captured and a position where the observation image is captured is estimated,
The billing processing unit
Based on the estimation result, charging processing is performed at the charging position corresponding to the reference image.
The vehicle-mounted device according to claim 8 or 9. The data processing unit
Based on the reference image and the observation image, obtain a predicted passage time when the user vehicle passes through the charging position corresponding to the reference image,
The billing processing unit
The user vehicle is charged at the estimated passage time.
The vehicle-mounted device according to any one of claims 8 to 11. The data processing unit
The estimated passage time is obtained based on a distance to the charging position calculated based on a relative positional relationship between the position where the reference image corresponding to the charging position is captured and the position where the observation image is captured. ,
The vehicle-mounted device according to claim 12. The data processing unit
Obtaining the predicted passage time based on a relative positional relationship between a position where the reference image corresponding to the charging position is photographed and a position where the observation image is photographed, and a traveling speed of the user vehicle;
The vehicle-mounted device according to claim 12 or 13.

Images