JP2020204709A - Map information update system and map information update method - Google Patents

Abstract

To provide a map information update system and a map information update method capable of updating a map reflecting the latest situation almost every day.SOLUTION: A map information update system acquires, at any time, a delivery ledger, in which address information of a delivery is updated sequentially in units of house number, sequentially acquires delivery image data in time series together with at least any of photographing position information and the related data at the time of delivery by a plurality of delivery vehicles for deliveries, collects the acquired delivery image data together with its photographing position information, refers to the acquired photographing position information regarding each piece of the collected delivery image data, detects difference on the images with the delivery image data previously photographed at the same photographing position, refers to the acquired delivery ledger, verifies matching of the detected difference on the images with address information at a house number corresponding to the photographing position where the difference is detected, and updates information based on the difference on the images at the photographing position with respect to the map information when the matching is verified.SELECTED DRAWING: Figure 5

Description

The present invention relates to a map information updating system and a map information updating method for sequentially updating existing map information based on delivery image data taken by a vehicle traveling on a road.

In recent years, applications that display a map on a screen and execute various operations via a PC, a smartphone, a tablet terminal, or the like have become widespread. Particularly in recent years, a so-called street view service has also been realized, which reproduces panoramic images of the entire circumference taken at various points on the road.

The map information displayed through such an application is acquired by performing image recognition processing on the image data taken about the actual road. In the past, it was common to use such image data captured through aerial photographs, but in recent years, in particular, there have been cases where traveling images taken from vehicles traveling on the road are used. It is increasing (see, for example, Patent Documents 1 and 2).

JP-A-2005-098853 JP-A-2010-156815

The disclosed technologies of Patent Documents 1 and 2 described above are all based on the premise that a high-performance omnidirectional shooting camera is mounted on a vehicle to create a highly accurate map. Such omnidirectional shooting cameras are generally expensive and can only be installed in a limited number of vehicles. For this reason, it takes a long time for a limited number of vehicles to travel on all the roads on the map. As a result, when focusing on a shooting point on a certain road, the running frequency of the limited vehicle equipped with the omnidirectional camera becomes extremely low, and the update interval of the map based on the shot running image becomes very low. There is a problem that it will be vacant for a long time.

If this update interval is vacant for a long period of time, there is a problem that if the road, building structure, civil engineering structure, or resident changes during that period, it will be delayed to be reflected on the map. Occurs. For example, even if a new road is opened, a new building is constructed, or a convenience store moves in as a new tenant of the building, this is the latest map. It will not be reflected.

Therefore, considering the convenience of users who use maps, it was necessary to propose a system that can utilize maps that reflect the latest situation almost every day.

Therefore, the map information updating system and the map information updating method to which the present invention is applied have been devised in order to solve the above-mentioned problems, and the purpose thereof is to take a picture with a vehicle traveling on the road. In the map information update system and the map information update method that sequentially update the existing map information based on at least one of the delivered delivery image data and the related data regarding the shooting position information and the delivery status, particularly the latest roads and building structures. The purpose is to provide a map information update system and a map information update method that can be updated almost every day on a map that reflects the situation of civil engineering structures, residents, etc.

In order to solve the above-mentioned problems, the map information updating system according to the present invention obtains a map information holding means for holding map information and a delivery ledger in which the destination information of deliveries is sequentially updated for each address. It is installed in each of the ledger information acquisition means and the delivery vehicle of a plurality of deliveries, and at the time of delivery of each delivery vehicle, the delivery image data is collected in chronological order together with at least one of the shooting position information and the related data regarding the delivery status. The data acquisition means to be sequentially acquired, the data collection server that collects the delivery image data acquired by each of the above data acquisition means together with the shooting position information, and each delivery image data collected by the data collection server are acquired. The difference detecting means for detecting the difference on the image with the delivery image data previously captured at the same shooting position by referring to the acquired position information, and the delivery ledger acquired by the delivery ledger information acquiring means. By referring to the consistency matching means for collating the difference on the image detected by the difference detecting means with the destination information at the address corresponding to the detected shooting position, and the consistency matching means. When the consistency is collated, the map information held by the map information holding means is provided with a map information updating means for updating the information based on the difference on the image at the shooting position. ..

In order to solve the above-mentioned problems, the map information updating system according to the present invention obtains a map information holding means for holding map information and a delivery ledger in which the destination information of deliveries is sequentially updated for each address. It is mounted on each of the ledger information acquisition means and the delivery vehicle of a plurality of deliveries, and at the time of delivery of each delivery vehicle, the delivery image data is sequentially acquired in chronological order together with at least one of the shooting position information and the above-mentioned related data. With respect to the data acquisition means to be performed and each delivery image data acquired by the data acquisition means, the delivery image previously captured at the same shooting position by referring to at least one of the acquired shooting position information and the related data. The difference detecting means for detecting the difference on the image with the data and the difference on the image of the delivery image data acquired by each of the above difference detecting means are collected together with the shooting position information and at least one of the related data. With reference to the data collection server to be used and the delivery ledger acquired by the delivery ledger information acquisition means, the difference on the image collected by the data collecting server and the destination information at the address corresponding to the detected shooting position. When the consistency is collated by the consistency collating means for collating the consistency with the above-mentioned consistency collating means, the map information held by the map information holding means is displayed on the image at the shooting position. It is characterized by providing a map information updating means for updating information based on the difference.

In order to solve the above-mentioned problems, the map information updating method according to the present invention includes a map information holding step for holding map information and a delivery for which the destination information of the delivery is sequentially updated for each address. The ledger information acquisition step, the data acquisition step of sequentially acquiring delivery image data together with the shooting position information at the time of delivery of a plurality of delivery vehicles, and the delivery image data acquired in each of the above data acquisition steps. Refer to at least one of the acquired shooting position information and the related data for each delivery image data collected in the data collection step and the data collection step of collecting the image together with at least one of the shooting position information and the related data. Then, referring to the difference detection step for detecting the difference on the image with the delivery image data previously captured at the same shooting position and the delivery ledger acquired in the delivery ledger information acquisition step, the difference detection step When the consistency is collated by the consistency collation step for collating the difference on the image detected by and the destination information at the address corresponding to the detected shooting position and the consistency collation step described above. Is characterized in that it has a map information update step for updating information based on a difference on an image at the shooting position with respect to the map information held in the map information holding step.

In order to solve the above-mentioned problems, the map information updating method according to the present invention includes a map information holding step for holding map information and a delivery for which the destination information of the delivery is sequentially updated for each address. The ledger information acquisition step, the data acquisition step of sequentially acquiring the delivery image data together with at least one of the shooting position information and the related data related to the delivery status at the time of delivery of the delivery vehicle of a plurality of deliveries, and the above data. For each delivery image data acquired in the acquisition step, the difference on the image is different from the delivery image data previously captured at the same shooting position by referring to at least one of the acquired shooting position information and the related data. The difference detection step for detecting the difference, the data collection step for collecting the difference on the image of the delivery image data acquired in each of the above difference detection steps together with the shooting position information, and the delivery ledger acquired in the above delivery ledger information acquisition step. Refer to the consistency collation step that collates the difference on the image collected in the data collection step with the destination information at the address corresponding to the detected shooting position, and the consistency collation step. When the sexes are collated, the map information held in the map information holding step is characterized by having a map information updating step for updating the information based on the difference on the image at the shooting position.

The map information updating system to which the present invention having the above-described configuration is applied captures delivery image data from a delivery vehicle that patrols the same location on the road on a daily basis. At least one of the position information and the related data related to the delivery status. By sequentially acquiring the data in chronological order and confirming the difference between the previous delivery image data captured at the same shooting position and the image, the update of the destination information at the shooting position is first grasped from the image. In addition, the update of the destination information is confirmed from the delivery ledger, not just the judgment on the image. This delivery ledger is also updated almost every day, and especially in the case of address change, relocation, or new establishment, that fact is first reflected in the destination information.

In other words, you can check the update by acquiring the delivery image data that reflects the latest situation on a daily basis, the shooting position information, the related data related to the delivery status, and the delivery ledger, and also check the update. If it is possible, the updated contents can be updated with respect to the map information.

Therefore, the map information update system to which the present invention is applied can reflect the latest road, building structure, civil engineering structure, resident, etc., in the map information, and the latest state is updated almost every day. It will be possible to update over.

It is a block diagram which shows the whole structure of the map information update system to which this invention is applied. It is a figure which shows the configuration example of the vehicle traveling on the road. It is a detailed block block diagram of a control device and a data acquisition part. It is a block block diagram around a data collection server. It is a figure for demonstrating an example of acquiring the photographing position information with respect to the delivery image imaged at a certain timing at the time of delivery. It is a figure which shows the example which detects the difference on the image between the two through the brightness of a pixel, etc. by the difference detection part. It is a figure for demonstrating the example of updating the map information by the map information update part. It is a figure which shows the example which acquires the combination association degree of 3 or more stages of the difference on an image, and the update of the map information with respect to the combination of acquired destination information in advance.

Hereinafter, the map information updating system to which the present invention is applied will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing an overall configuration of a map information updating system 1 to which the present invention is applied. The map information update system 1 is provided for each regional base, and is a relay server 10 that collects delivery image data captured in the area of the collection and delivery station under its umbrella together with at least one of shooting position information and related data related to the delivery status. And a data collection server 11 provided at the center, which collects the delivery image data collected in each relay server 10 together with at least one of the shooting position information and the related data regarding the delivery status at one place. Delivery image data can be transmitted and received between the servers 10 and 11 together with at least one of shooting position information and related data related to the delivery status by wired communication or wireless communication.

Further, the relay server 10 makes it possible to receive the delivery image data together with at least one of the shooting position information and the related data regarding the delivery status by wireless communication with the radio base stations 12 spread in various places in the area under the jurisdiction. ..

As shown in FIG. 2, the radio base station 12 receives the delivery image data from the vehicle 14 traveling on the road together with at least one of the shooting position information and the related data regarding the delivery status.

The vehicle 14 also includes a four-wheeled vehicle such as an automobile or a truck, a two-wheeled vehicle such as a motorcycle or a motorcycle, and the like. The vehicle 14 is a so-called delivery vehicle for delivering deliverables. The vehicle 14 includes a control device 2 for acquiring a delivery image together with at least one of shooting position information and related data related to the delivery status, and a data acquisition unit 3 for acquiring the shooting position by the control device 2.

FIG. 3 shows a detailed block configuration diagram of the control device 2 and the data acquisition unit 3.

The control device 2 includes a control unit 21, an acquisition unit 22, an operation unit 23, a communication unit 24, and a storage unit 25 connected to the control unit 21, and further, the control unit 21 includes a data acquisition unit 3 Is connected.

The control unit 21 has a CPU (Central Processing Unit) 211 and a memory 212 connected to the CPU 211, respectively.

The memory 212 is a storage means embodied as a ROM (Read Only Memory) or a RAM (Random Access Memory). The ROM stores a program for controlling the hardware resources of the entire control device 2. The RAM is also used as a work area used for data storage and expansion, and temporarily stores various instructions for controlling the hardware resources of the entire control device 2.

The CPU 211 is a so-called central arithmetic unit for controlling all the components. The CPU 211 notifies each component of an instruction for reading a program stored in the memory 212 and performing various operations. For example, if the program stored in the memory 212 captures the delivery image data in the 360 ° omnidirectional direction at regular time intervals, the acquisition unit 22 is executed based on this. Send a signal for Further, when the program stored in the memory 212 acquires the current position information or the related data regarding the delivery status at the same time as the acquisition unit 22 captures the delivery image data, the acquisition unit 22 and the data acquisition unit 3 are contacted. A signal is transmitted to execute such an instruction.

Further, the CPU 211 controls the delivery image data acquired by the acquisition unit 22 and temporarily stored in the storage unit 25 to be transmitted via the communication unit 24 together with at least one of the shooting position information and the related data related to the delivery status. To execute.

The acquisition unit 22 is composed of, for example, a lens (not shown) and an image sensor, and is composed of a camera capable of acquiring an image in all directions over a 360 ° direction. In addition, the acquisition unit 22 acquires various related data regarding the delivery status. The related data may include, for example, text data, signals, voice data, as well as image data of deliveries and weather. The timing of imaging or acquisition by the acquisition unit 22 is controlled by the CPU 211 in the control unit 21. The omnidirectional delivery image data captured by the acquisition unit 22 is stored once in the storage unit 25 together with at least one of the shooting position information and the related data regarding the delivery status. Incidentally, the acquisition unit 22 is not limited to the case of acquiring an omnidirectional image, but also includes the case of acquiring an image of only one position.

The operation unit 23 is composed of a touch panel, buttons, levers, and the like for inputting operation information for operating the control device 2. The operation unit 23 also includes a liquid crystal panel and the like for displaying various information to the user. The operation unit 23 sends the input operation information to the control unit 21. The control unit 21 performs various controls based on the operation information transmitted from the operation unit 23.

Under the control of the control unit 21, the communication unit 24 transmits the delivery image data captured by the acquisition unit 22 to the wireless base station 12 via wireless communication together with at least one of the shooting position information and the related data regarding the delivery status. The communication unit 24 also includes an antenna that performs frequency conversion and other various conversion processes necessary for performing wireless communication, converts an electric signal into a radio wave, or converts a radio wave into an electric signal. The communication unit 24 also has a function of receiving a signal from the radio base station 12, and when receiving a signal from the radio base station 12, sends the signal to the control unit 21.

The storage unit 25 is composed of a hard disk, a memory, or the like that can store delivery image data together with at least one of shooting position information and related data related to the delivery status. The delivery image data is stored in the storage unit 25 together with at least one of the shooting position information and the related data regarding the delivery status under the control of the control unit 21, and is stored under the control of the control unit 21. The delivery image data will be read out together with at least one of the shooting position information and the related data regarding the delivery status.

The data acquisition unit 3 acquires the current position information of the vehicle 14 in real time based on the satellite positioning signal sent from the artificial satellite. In the process of delivering the vehicle 14 on the road, the data acquisition unit 3 receives the satellite positioning signal at any time, so that the position information at each point on the delivered road can be acquired. Further, by receiving the satellite positioning signal by the data acquisition unit 3 at the timing of capturing the image by the acquisition unit 22, it is possible to acquire the position information at the imaging point of the image. When the data acquisition unit 3 actually receives the sanitary positioning signal, and by extension, when the position information is acquired, the control unit 21 controls it. Therefore, by controlling the acquisition timing of the satellite positioning signal by the data acquisition unit 3 and the imaging timing by the acquisition unit 22 at the same time through the control unit 21, it is possible to acquire the position information at the imaging point. ..

FIG. 4 shows a block configuration around the data collection server 11. The data collection server 11 includes a map information holding unit 31, a delivery ledger information acquisition unit 32, a difference detection unit 33, a consistency collation unit 34, and a map information update unit 35.

The map information holding unit 31 stores map information including a map of Japan and a map of each country in the world. The map information referred to here is embodied in a two-dimensional map that describes the map in a plane, a three-dimensional map that describes the map in three dimensions, and a street view image consisting of a panoramic image of the entire circumference taken at a point on the road. It is configured as electronic data. Based on such map information, it is possible to display a map on the screen via a PC, smartphone, tablet terminal, etc., and by extension, it is possible to execute various operations on this displayed map through an application. It becomes. The map information holding unit 31 may acquire the electronic data of the map posted on the Internet from the initial map information, or the electronic data of the map distributed free of charge or commercially available. May be obtained. The map information obtained and held by the map information holding unit 31 in this way will be updated at any time by the map information updating unit 35, which will be described later.

The delivery ledger information acquisition unit 32 acquires a delivery ledger at any time, in which the destination information of the deliveries is sequentially updated for each street address. This delivery ledger is prepared by the post office and contains a list of residents of all residents in the area by street address. In the address information in this delivery ledger, the householder's name, the company name, etc. are described with respect to the street number and the room number. In the address information, not only the names of the residents of all the rooms of the apartment house but also the names of the residents of all the rooms of the apartment house are listed in units of room numbers. The destination information in this delivery ledger allows the delivery person to confirm the current residence of the addressee by comparing it with the delivery within the station before the delivery departs. In addition to such residents, the destination information in the delivery ledger also includes information on companies and stores that are occupying as tenants. Further, in this destination information, information about a building, a condominium, a detached house, etc. is newly described in the destination information.

The address information in this delivery ledger is usually reflected immediately when there is a change in the address of residents, tenants, etc., and if a new address is created due to the construction of a new road, etc., this is also immediately registered. In addition, any changes in the content due to new construction or repair of building structures will be reflected immediately. The delivery ledger information acquisition unit 32 acquires such a delivery ledger every day or once every few days. Therefore, the delivery ledger information acquisition unit 32 can acquire the destination information with the latest status updated.

The difference detection unit 33 detects a difference on the image of at least one of the delivery image data and the related data collected by the data collection server from the delivery image data previously captured at the same shooting position. .. This difference on the image is performed on a pixel-by-pixel basis or on a block-by-block basis composed of a plurality of pixels, and the presence or absence of the difference is determined based on the brightness and the like by applying the conventional image processing technique.

The consistency collation unit 34 collates the consistency of the difference on the image detected by the difference detection unit 33 with the destination information at the address corresponding to the shooting position where the difference is detected. Consistency with this destination information is performed by referring to the delivery ledger acquired by the delivery ledger information acquisition unit 32. When the consistency is collated, the consistency collation unit 34 notifies the map information update unit 35 to that effect.

When the consistency is collated by the consistency collation unit 34, the map information updating unit 35 updates the information based on the difference on the image with respect to the map information held by the map information holding unit 31.

Next, an operation example of the map information updating system 1 to which the present invention is applied will be described.

It is premised that the vehicle 14 in which the control device 2 and the data acquisition unit 3 are installed is a delivery vehicle for delivering the delivered product as described above. For this reason, the vehicle 14 travels almost every day in various places within its jurisdiction in order to deliver the deliverables. In the process in which the vehicle 14 travels, the acquisition unit 22 of the control device 2 captures a delivery image consisting of all directions, and the data acquisition unit 3 acquires at least one of the position information and the related data. The control unit 21 controls the imaging timing of the delivered image and the acquisition timing of at least one of the position information and the related data by the data acquisition unit 3 at the same time. As a result, at least one of the position information and the related data acquired by the data acquisition unit 3 corresponds to the time when the delivery image is captured. Hereinafter, the position information corresponding to the time when the delivered image is captured is referred to as the shooting position information.

Such acquisition of the delivery image by the acquisition unit 22 and acquisition of at least one of the imaged position information and the related data regarding the delivery status by the data acquisition unit 3 are continuously executed. As a result, as shown in FIG. 5, at least one of the shooting position information (x1, y1) and the related data Q1 can be acquired for the delivery image P1 captured at a certain timing t1 at the time of delivery. .. At least one of the shooting position information (x2, y2) and the related data Q2 can be acquired for the delivery image P2 captured at the next timing t2. Similarly, at least one of the shooting position information (xn, yn) and the related data Qn can be acquired for the delivery image Pn captured at the next timing tn.

As related data regarding the delivery status, the acquisition unit 22 obtains data including at least one of, for example, time, distance, route, delivery object, delivery person, store, facility, transportation, weather, temperature, humidity, and environmental information. get. The related data may be, for example, text data, voice data, signal data, and image data.

The control device 2 transmits at least one of the delivery image Pn captured at each timing tn, the shooting position information (xn, yn), and the related data Qn to the outside via the communication unit 24, respectively. .. At least one of the delivery image Pn, the shooting position information (xn, yn), and the related data Qn is sent to the data collection server 11 via the relay server 10. As a result, the control device 2 is sent at least one of the delivery image Pn, the acquired position information (xn, yn), and the related data Qn in chronological order from the vehicle 14 to the timing t1 to the timing tn. Will come. As a result, the delivery images P1 to Pn in at least one of the acquired position information (x1, y1) to (xn, yn) and the related data Qn are accumulated in the data collection server 11.

Further, at least one of the delivery image Pn, the acquired position information (xn, yn), and the related data Qn captured by the control devices 2 installed in all the vehicles 14 is sent to the data collection server 11. As a result, if each vehicle 14 patrols within its own jurisdiction, at least one of the delivery image Pn, the acquired position information (xn, yn), and the related data Qn captured from the roads of all the jurisdictions. Will be sequentially accumulated in the data collection server 11 in chronological order.

The data collection server 11 detects at least one of the delivery image Pn, the acquired position information (xn, yn), and the related data Qn accumulated in this way via the difference detection unit 33, which will be described below. Perform the processing operation for.

The difference detection unit 33 detects the difference on the image of the newly acquired delivery image data from the delivery image data previously captured at the same shooting position. In the case of the above-mentioned example of FIG. 5, when the delivery image data P2 captured at the timing t2 is newly acquired, it is previously captured at at least one of the same shooting position (x2, y2) and the related data Q2. Read the delivered image data. If the delivery image data P2'captured at the same shooting position (x2, y2) and at least one of the related data Q2 on the previous day is acquired, the delivery image data P2 and the delivery image data P2' Detects differences on the image with. By the way, the identity of the shooting position between the delivery image data P2 and the delivery image data P2'is determined based on the shooting position information associated with each of them. The difference detection unit 33 may detect the difference on the image by using a deep learning technique.

When the difference detection unit 33 detects the difference on the image between the two through, for example, the brightness of the pixels, for example, as shown in FIG. 6, the delivery image data P2 has a new signboard of "○ × Shoji". It is assumed that the listed points are detected as differences on the image. The delivery image data P2'captured the day before did not have this "○ × Shoji" sign, so this was detected as a difference on the image. In other words, this suggests that a new "○ × Shoji" sign was put up from the day before to today.

The difference detection unit 33 refers to at least one of the acquired shooting position information and the related data Qn for each delivery image data Pn collected by the data collection server 11. Then, the difference detection unit 33 narrows down the delivery image data Pn to be detected based on, for example, the related data Qn, and detects the difference on the image between the delivery image data Pn previously captured at the same shooting position. ..

On the image based on the change between the delivery image data P2 and the delivery image data P2', which have different imaging time points at least one of the same acquisition position (x2, y2) and the related data Q2. When the difference is detected by the difference detection unit 33, the consistency collation unit 34 refers to the delivery ledger acquired by the delivery ledger information acquisition unit 32. Since the delivery address information is updated almost every day in the delivery ledger information acquisition unit 32 for each street address, the latest information about the resident can be acquired by accessing this. The consistency collation unit 34 accesses the delivery ledger information acquisition unit 32 and confirms the destination information of the address corresponding to at least one of the shooting position (x2, y2) and the related data Q2. From the viewpoint of improving the convenience of actually confirming the address information of the address with respect to the shooting position, a correspondence table between the address and the shooting position information may be provided in the delivery ledger information acquisition unit 32. In this correspondence table, for example, if the shooting position information is represented by latitude and longitude, the latitude and longitude may be assigned to each address.

Further, the difference detection unit 3 displays, for example, a difference on the image between the delivery image data P2 and the delivery image data P2'based on the change between the two, such as a road sign, a road marking, a road marking, a guide display, and a construction. Detection may be performed using an image consisting of at least one of a structure, a civil engineering structure, a signboard, a guide board, a gathering post, a mailbox, a delivery box, a person, a resident, and an entrance / exit. This makes it possible to detect a difference in a specific part or range included in the delivery image data.

As a result of confirming the destination information, the consistency collation unit 34 newly finds that the address corresponding to the shooting position is "○ × Shosha", that the resident has moved to that address, which is a difference on the image. In addition, it can be seen from the delivery ledger information that it was confirmed. That is, it can be seen that the consistency on the image detected by the difference detection unit 33 and the destination information at the address corresponding to the detected shooting position are collated.

In such a case, the map information updating unit 35 indicates the resident of the address (shooting position x2, y2) with respect to the map information held by the map information holding unit 31 as shown in FIG. Update to.

From the viewpoint of enhancing the convenience of consistency confirmation, the consistency collation unit 34 accesses the delivery ledger information acquisition unit 32 at any time, and when it is newly updated, acquires the updated information. You may try to keep it.

In addition, the matching system collation unit 34 includes, for example, road markings, road markings, road markings, guidance displays, building structures, civil engineering structures, signboards, information boards, collective posts, mailboxes, delivery boxes, people, residents, and doorways. The consistency on the image based on the change with respect to at least one of the above may be collated with the above-mentioned destination information reflecting the change. This makes it possible to collate the consistency of a specific part or range included in the delivered image data.

Further, the delivery ledger information acquisition unit 32 may accumulate past delivery ledger information in chronological order. As a result, it is possible to read and confirm whether or not the address corresponding to the shooting position has been updated to "○ × Shoji" as the destination information of the delivery ledger as in this case. It is possible to improve the convenience of the collation work described above.

Further, according to the present invention, when character information such as a signboard is reflected in such a difference on an image, the character information is used as a character string by utilizing deep learning technology or a well-known character recognition technology. You may try to recognize the contents. Then, the consistency between the recognized character string and the destination information in the delivery ledger may be confirmed, and the identity may be confirmed.

By the way, the difference on the image is when a building structure such as a building or a house is newly constructed or a civil engineering structure such as a tunnel or a bridge is newly completed in addition to the signboard reflecting such text information. Is the same. For example, if a building that was under construction is newly completed and can be moved in, the difference in the image will be that it will appear between the building before the scaffolding etc. was removed and the completed building where the scaffolding was removed. become. Even when such a difference on the image can be detected, the consistency collation unit 34 similarly accesses the delivery ledger information acquisition unit 32 and confirms the address information of the address corresponding to the shooting position of the building. To do. When the building is completed and it becomes possible to move in, it will naturally be updated in the delivery ledger, so by comparing it with this, it is possible to confirm the consistency of the destination information and confirm the identity. Become.

The same applies when a new road is constructed, and the difference on the image can be detected between the image before the road is completed and the image when the road is completed. In such a case, the consistency collation unit 34 accesses the delivery ledger information acquisition unit 32 and confirms the destination information of the address corresponding to the road. If the address where the road is newly established is updated, that fact will naturally be updated in the delivery ledger, so it is possible to confirm the consistency and identity of the destination information by comparing it with this. It will be possible.

After the collation by the consistency collation unit 34 is completed in this way, the map information update unit 35 sequentially updates the map information held by the map information holding unit 31.

The map information updating system 1 to which the present invention is applied captures such delivery image data from a delivery vehicle that patrols the same location on the road on a daily basis, and obtains related data regarding the same shooting position and delivery status. By confirming the difference on the image from the previous running image data captured in at least one of them, the update of the destination information in at least one of the related data regarding the shooting position and the delivery status is first grasped from the image. In addition, the update of the destination information is confirmed from the delivery ledger, not just the judgment on the image. This delivery ledger is also updated almost every day, and especially in the case of address change, relocation, or new establishment, that fact is first reflected in the destination information.

In other words, you can get the delivery image data that reflects the latest situation and the destination information of the delivery ledger on a daily basis and check the update, and if you can check the update, The updated contents can be updated with respect to the map information.

Therefore, the map information update system 1 to which the present invention is applied can reflect the latest road, building structure, civil engineering structure, resident, etc. in the map information, and the latest state is updated almost every day. It is possible to update over.

The map information updating system 1 to which the present invention is applied updates not only the two-dimensional map as shown in FIG. 7 described above but also the three-dimensional map and the street view image when updating the map information. There may be. Since the delivery image data captures an image in the 360 ° direction in all directions, it is possible to acquire a three-dimensional shape of a building structure or a civil engineering structure, and this can be reflected on a map. For example, when it is confirmed that the building structure shown in the delivery image data P2 of FIG. 6 has been newly constructed, the shape of the building structure is drawn on the three-dimensional map when updating the map information. You may. The three-dimensional shape of the building structure will be drawn based on the image of the captured delivery image data.

Similarly, when updating the street view, for example, when the signboard "○ × Shoji" is newly updated in the delivery image data P2 of FIG. 6, the delivery image data P2 reflecting the signboard “○ × Shoji” is used as it is. You may update it as Street View.

In the above-described embodiment, the delivery image data captured in the control device 2 is directly transmitted to the data collection server 11 via the communication unit 24 together with at least one of the shooting position information and the related data regarding the delivery status. The explanation was given by taking the case as an example, but the explanation is not limited to this. In the control device 2, the delivery image data previously captured at the same shooting position may be detected as a difference on the image between at least one of the shooting position information and the related data regarding the delivery status. In such a case, a configuration corresponding to the difference detection unit 33 described above is provided in the control device 2. Then, each time the delivery image data is captured, the shooting position information is referred to, and the delivery image data previously captured at the same shooting position is read out and the difference on the image is compared. When a difference on the image is detected, the detected difference on the image is transmitted to the data collection server 11 via the communication unit 24.

As a result, if only the difference on the image and the shooting position information are transmitted via the communication unit 24, the amount of data transmitted can be significantly reduced, and the efficiency of the entire system can be improved. it can.

The data collection server 11 receives the difference on the image, confirms the destination information of the address corresponding to the shooting position, and updates the map information.

By the way, according to the present invention, the delivery image data is acquired every day at the time of delivery of each delivery vehicle, and the delivery ledger acquired every day is referred to to obtain the acquisition date of the delivery image data and the update date of the delivery ledger. It is premised that the matching dates are collated after matching, but the present invention is not limited to this, and the delivery image data and the delivery ledger may be acquired every few days.

Further, the map information updating system 1 to which the present invention is applied may execute the actual consistency determination based on the scheme described below.

In this scheme, as shown in FIG. 8, the degree of combination association of three or more stages of the difference on the image and the update of the map information with respect to the acquired combination of destination information is acquired in advance. This combination association degree is an index showing how strongly the difference on the image and the update of the map information with respect to the acquired destination information are related.

The update of the map information referred to here indicates what kind of update is actually reflected in the map information. For example, the update c is a change of the resident to a specific address in the map information. It may be set to add an update, or update d may be set to update a building name set at a specific address in the map information. In addition, update e may indicate that a new road has been constructed and as a result, a new address has been set.

The combination linkage degree shown in FIG. 8 means that a set of combinations of one or more differences on the image and one or more acquired destination information is expressed as so-called hidden layer nodes 61a to 61e. Each of the nodes 61a to 61e is set with a weighting for the difference on the image, the acquired destination information, and a weighting for updating the map information as an output solution.

This weighting is a combination linkage degree of three or more stages. For example, in the node 61a, the difference P51 on a certain image is associated with a combination association degree of 80%, and the acquired destination information: address change is associated with 80%. In the node 61c, the difference P52 on the image is associated with a combination association degree of 60%, the acquired destination information: no update is associated with a combination association degree of 60%, and the acquired destination information: resident change is associated with a combination association degree of 40%. ing. The degree of combination linkage may be configured by a neural network.

Such combination association degree is acquired in advance. Then, when trying to update the map information, the difference on the image constituting the combination and the destination information are acquired.

The search program performs an operation of selecting one or two or more updates c to g of map information as the optimum solution from the output solutions based on the acquired difference on the image and the destination information. In selecting this optimum solution, the combination association degree shown in FIG. 8 acquired in advance is referred to. For example, if the difference on the acquired image is P51 and the acquired destination information is the update of the building name, the node 61b is associated through the combination association degree, and the update e is the combination association of the node 61b. The degree 60% and the update g are associated with a combination association degree 40%. The optimum solution is searched for based on such combination linkage. It is not necessary to output the optimum solution having the highest degree of combination association, and the solution having the lower degree of combination association may be selected. Of course, other than this, an output solution to which the arrows are not connected may be selected. That is, the selection of the output solution is not limited to the case where the combination linkage degree is selected in order from the one having the highest combination linkage degree, and the combination linkage degree may be selected in order from the one having the lowest combination linkage degree depending on the case. However, it may be selected in any other priority.

As described above, according to the present invention, it is possible to efficiently search for a method of updating map information for which a new search is attempted. In particular, according to the present invention, it is possible to easily search for a method for updating map information without requiring special skill. Further, according to the present invention, it is possible to search for a method for updating map information with higher accuracy. Further, by constructing the above-mentioned combination linkage degree by artificial intelligence, it is possible to further improve the discrimination accuracy by learning this.

Further, the present invention is characterized in that a patent or a group of patents is searched for through a combination linkage degree set to three or more stages. The degree of combination linkage can be described by a numerical value from 0 to 100%, for example, but is not limited to this, and may be configured at any stage as long as it can be described by a numerical value of 3 or more stages. ..

By searching based on the combination association degree represented by the numerical values of three or more stages, the search is performed in descending order of the combination association degree in the situation where the determination result of the method for updating a plurality of map information is selected. It is also possible to display it. If the user can be displayed in descending order of the degree of combination association in this way, it is possible to prompt the user to preferentially select the determination result of the more probable update method of the map information. On the other hand, even the determination result of the update method of the map information having a low degree of combination association can be displayed in the meaning of the second opinion, and can be useful when the analysis cannot be performed well by the first opinion.

In addition to this, according to the present invention, it is possible to determine without overlooking the determination result of the update method of the map information having an extremely low degree of combination association such as 1%. Even patents or patent groups with extremely low degree of combination association are connected as a slight sign, and may be useful as a discrimination result of patents or patent groups once every tens or hundreds of times. It is possible to alert the user.

Further, according to the present invention, there is an advantage that the search policy can be determined by the method of setting the threshold value by performing the search based on the combination linkage degree of three or more stages. If the threshold value is lowered, even if the above-mentioned combination association degree is 1%, it can be picked up without omission, but it is unlikely that the determination result of the map information update method can be detected favorably, and a lot of noise is picked up. It may end up. On the other hand, if the threshold value is raised, there is a high possibility that the determination result of the optimum map information update method can be detected with high probability, but the combination linkage degree is usually low and it is passed through, but it is tens or hundreds of times. In some cases, the suitable solution that comes out once is overlooked. It is possible to decide which one should be emphasized based on the ideas of the user side and the system side, but it is possible to increase the degree of freedom in selecting the points to be emphasized.

Further, the degree of association between these combinations may be learned by the artificial intelligence. Similarly, the degree of association with the patent or the patent group for the combination with the patent specific information and the incidental information may be learned by artificial intelligence.

As a result of applying the present invention, the difference on the image with respect to the update of the map information selected by the user or the system and the relationship with the acquired destination information are acquired, and the above-mentioned combination linkage degree is updated based on this. That is, relationships selected by the user or system are updated with high relevance, and relationships not selected by the user or system are updated with low relevance. For this update, artificial intelligence, neural networks, or the like may be used. This update may be performed periodically or irregularly each time a new difference on the image is acquired.

Although the embodiment of the present invention has been described above, this embodiment is presented as an example and is not intended to limit the scope of the invention. These novel embodiments can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are also included in the scope of the invention described in the claims and the equivalent scope thereof.

1 Map information update system 2 Control device 3 Data acquisition unit 10 Relay server 11 Data collection server 12 Wireless base station 14 Vehicle 21 Control unit 22 Acquisition unit 23 Operation unit 24 Communication unit 25 Storage unit 31 Map information holding unit 32 Delivery ledger information acquisition Unit 33 Difference detection unit 34 Consistency collation unit 35 Map information update unit 211 CPU
212 memory

Claims (16)

Map information holding means for holding map information and
Delivery ledger information acquisition means for acquiring delivery ledger at any time, in which the destination information of deliveries is updated sequentially for each street address.
A data acquisition means that is installed in each of a plurality of delivery vehicles and sequentially acquires delivery image data together with at least one of shooting position information and related data related to the delivery status at the time of delivery of each delivery vehicle. When,
A data collection server that collects delivery image data acquired by each of the above data acquisition means together with at least one of the shooting position information and the related data.
For each delivery image data collected by the data collection server, at least one of the acquired shooting position information and the related data is referred to, and an image is displayed between the delivery image data and the delivery image data previously captured at the same shooting position. Difference detection means to detect the above difference and
The delivery ledger acquired by the delivery ledger information acquisition means is referred to, and the consistency of the difference on the image detected by the difference detecting means and the destination information at the address corresponding to the detected shooting position is collated. Consistency collation means and
When the consistency is collated by the above-mentioned consistency collating means, the map information updating means for updating the information based on the difference on the image at the shooting position with respect to the map information held by the map information holding means. A map information update system characterized by being prepared. Map information holding means for holding map information and
Delivery ledger information acquisition means for acquiring delivery ledger at any time, in which the destination information of deliveries is updated sequentially for each street address.
A data acquisition means that is installed in each of a plurality of delivery vehicles and sequentially acquires delivery image data together with at least one of shooting position information and related data related to the delivery status at the time of delivery of each delivery vehicle. When,
For each delivery image data acquired by the data acquisition means, refer to at least one of the acquired shooting position information and the related data, and image the image with the delivery image data previously captured at the same shooting position. Difference detection means to detect the above difference and
A data collection server that collects the difference on the image of the delivery image data acquired by each of the above difference detecting means together with at least one of the shooting position information and the related data.
The delivery ledger acquired by the delivery ledger information acquisition means is referred to, and the consistency of the difference on the image collected by the data collecting server with the destination information at the address corresponding to the detected shooting position is collated. Consistency collation means and
When the consistency is collated by the above-mentioned consistency collating means, the map information updating means for updating the information based on the difference on the image at the shooting position with respect to the map information held by the map information holding means. A map information update system characterized by being prepared. The data acquisition means obtains the delivery status of the related data as data consisting of at least one of time, distance, route, delivery object, delivery person, store, facility, transportation, weather, temperature, humidity, and environmental information. The map information update system according to claim 1 or 2, characterized in that the data is acquired by. The difference detecting means detects the difference in the image as a road sign, a road marking, a road marking line, a guide display, a building structure, a civil engineering structure, a signboard, a guide board, a gathering post, a mailbox, a delivery box, a person, a resident. The map information updating system according to claim 1 or 2, wherein the detection is performed using an image consisting of at least one of the entrance and exit. The map information updating system according to claim 1 or 2, wherein the difference detecting means detects a difference on the image by using a deep learning technique. The above consistency verification means are at least road signs, road markings, road markings, guidance signs, building structures, civil engineering structures, signs, information boards, collective posts, mailboxes, parcel boxes, people, residents, and doorways. The map information updating system according to any one of claims 1 to 5, wherein the difference on the image based on any one of the changes is collated with the above-mentioned destination information reflecting the change. .. The above data acquisition means acquires delivery image data every day at the time of delivery of each delivery vehicle.
The consistency collation means refers to the delivery ledger acquired daily by the delivery ledger information acquisition means, and collates the consistency date after matching the acquisition date of the delivery image data and the update date of the delivery ledger. The map information updating system according to any one of claims 1 to 6, characterized in that. The map information holding means holds map information including a two-dimensional map, a three-dimensional map, or a street view image as a panoramic image taken at each shooting position.
The map information updating means is characterized in that the information based on the difference on the image at the shooting position is updated for any of the held two-dimensional map, three-dimensional map, and street view image. The map information update system according to any one of ~ 7. Map information retention step to retain map information and
The delivery ledger information acquisition step for acquiring the delivery ledger at any time, in which the delivery address information is sequentially updated for each street address, and
A data acquisition step of sequentially acquiring delivery image data together with at least one of shooting position information and related data related to the delivery status at the time of delivery of a plurality of delivery vehicles, and a data acquisition step.
A data collection step of collecting the delivery image data acquired in each of the above data acquisition steps together with at least one of the shooting position information and the above-mentioned related data.
For each delivery image data collected in the data collection step, at least one of the acquired shooting position information and the related data is referred to, and the delivery image data previously captured at the same shooting position is displayed on the image. Difference detection step to detect the difference in
The delivery ledger acquired in the delivery ledger information acquisition step is referred to, and the consistency of the difference on the image detected by the difference detection step with the destination information at the address corresponding to the detected shooting position is collated. Consistency collation step and
When the consistency is collated by the above-mentioned consistency collation step, it has a map information update step for updating the information based on the difference on the image at the shooting position with respect to the map information held in the map information retention step. A method of updating map information, which is characterized by this. Map information retention step to retain map information and
The delivery ledger information acquisition step for acquiring the delivery ledger at any time, in which the delivery address information is sequentially updated for each street address, and
A data acquisition step of sequentially acquiring delivery image data together with at least one of shooting position information and related data related to the delivery status at the time of delivery of a plurality of delivery vehicles, and a data acquisition step.
For each delivery image data acquired in the data acquisition step, at least one of the acquired shooting position information and the related data is referred to, and the image is displayed between the delivery image data previously captured at the same shooting position. Difference detection step to detect the difference in
A data collection step that collects the difference on the image of the delivery image data acquired in each of the above difference detection steps together with the shooting position information, and
Matching that refers to the delivery ledger acquired in the delivery ledger information acquisition step and collates the consistency of the difference on the image collected in the data collecting step with the destination information at the address corresponding to the detected shooting position. Gender matching steps and
When the consistency is collated by the above-mentioned consistency collation step, it has a map information update step for updating the information based on the difference on the image at the shooting position with respect to the map information held in the map information retention step. A method of updating map information, which is characterized by this. The data acquisition step sets the delivery status of the related data as data consisting of at least one of time, distance, route, delivery object, delivery person, store, facility, transportation, weather, temperature, humidity, and environmental information. The map information updating method according to claim 9 or 10, wherein the map information is acquired by. In the difference detection step, the difference in the above image is detected by road signs, road markings, road markings, guidance displays, building structures, civil engineering structures, signboards, information boards, collective posts, mailboxes, delivery boxes, people, and residents. The map information updating method according to claim 9 or 10, wherein the detection is performed using an image consisting of at least one of the entrance and exit. The map information updating method according to claim 9 or 10, wherein in the difference detection step, a difference on the image is detected by using a deep learning technique. In the above consistency verification step, at least of road signs, road markings, road markings, guidance signs, building structures, civil engineering structures, signboards, information boards, gathering posts, mailboxes, parcel boxes, people, residents, and doorways. The map information update method according to any one of claims 9 to 13, wherein the difference on the image based on any of the changes is collated with the above-mentioned destination information reflecting the change. .. In the above data acquisition step, delivery image data is acquired every day at the time of delivery of each delivery vehicle.
In the consistency collation step, the delivery ledger acquired every day in the delivery ledger information acquisition step is referred to, and the matching date is collated after matching the acquisition date of the delivery image data and the update date of the delivery ledger. The map information updating method according to any one of claims 9 to 14, wherein the map information is updated. In the map information holding step, map information consisting of a two-dimensional map, a three-dimensional map, or a street view image as a panoramic image taken at each shooting position is held.
9. The map information update step is characterized in that information based on the difference on the image at the shooting position is updated for any of the held two-dimensional map, three-dimensional map, and street view image. The map information updating method according to any one of ~ 15.

Images