WO2016031229A1

WO2016031229A1 - Road map creation system, data processing device, and on-board device

Info

Publication number: WO2016031229A1
Application number: PCT/JP2015/004248
Authority: WO
Inventors: 朋明阿部
Original assignee: パナソニックＩｐマネジメント株式会社
Priority date: 2014-08-27
Filing date: 2015-08-25
Publication date: 2016-03-03
Also published as: JPWO2016031229A1

Abstract

A road map creation system is provided with a plurality of on-board devices that are mounted in a plurality of vehicles and each of which transmits a captured image of the outside of each vehicle to a server device, and a data processing device that creates a road image by using a plurality of captured images collected in the server device. The data processing device includes an image search unit that searches for a plurality of common captured images each including an image of a common spot among the collected captured images, a feature extracting unit that extracts a feature within the plurality of collected captured images, and a high-definition processing unit that generates a higher-definition road image than the common captured images by combining the common captured images through association of pixels representing an identical part of a subject in the plurality of common captured images on the basis of the feature.

Description

Road map creation system, data processing device and in-vehicle device

The present invention relates to a road map creation system, a data processing device, and an in-vehicle device that create a high-definition road image.

A system for automatic driving of vehicles is being studied. In this system, for example, how to automatically steer a vehicle based on road map data, for example, by determining the curvature, slope, unevenness, and the exact position of the vehicle on the road, etc. Judge. For this reason, high-definition road map data is required.

High-definition road map data can be created by preparing a special road measurement vehicle and running the road measurement vehicle on an actual road to measure the road.

Conventionally, a system has been proposed that collects captured images from a plurality of vehicles or collects captured images from a plurality of users, and generates a three-dimensional map data by combining these collected captured images (for example, (See Patent Documents 1 and 2).

In addition, a system for creating and updating a stereoscopic image of a building or the like on a map by combining a photographed image with two-dimensional map data has been disclosed (for example, see Patent Document 3).

Japanese Patent Laid-Open No. 2002-189412 Japanese Patent No. 5472142 International Publication No. 2008/062819

A road map creation system according to the present disclosure is mounted on a plurality of vehicles, roads using a plurality of in-vehicle devices that transmit outdoor captured images of the vehicles to a server device, and a plurality of captured images collected by the server device. A data processing device for creating an image, and the data processing device searches for a plurality of common photographed images including a common point image from the collected photographed images; A feature extraction unit that extracts features in the collected images of the plurality of captured images, and the plurality of common captured images that are based on the features and the positions of the images are aligned. A high-definition processing unit that generates a road image having a higher definition than that of each common captured image from information of each pixel of the common captured image.

A data processing device of the present disclosure is a data processing device that creates a road image using a plurality of captured images that are captured by a plurality of vehicles and collected by a server device, and the data processing device of the collected plurality of captured images An image search unit that searches for a plurality of common captured images including images of a common spot, a feature extraction unit that extracts features in the collected images of the plurality of captured images, and the feature Based on the information of each pixel of the plurality of common captured images obtained by aligning the positions of the images based on the plurality of common captured images, a high-definition road image that is higher in definition than the common captured images is generated. And a processing unit.

The in-vehicle device according to the present disclosure includes: an image search unit that searches for a plurality of common captured images including a common spot image from a plurality of captured images collected by the server device; and the plurality of collected captured images And a feature extraction unit that extracts a feature in the image, and aligns the position of the image between the plurality of common captured images based on the feature, and information on each pixel of the plurality of common captured images A data processing device having a high-definition processing unit that generates a high-definition road image than each common captured image, and the vehicle-mounted device mounted on the vehicle, An input unit that inputs an outdoor captured image from the imaging device mounted on the camera, and a communication unit that transmits the outdoor captured image to the server device.

FIG. 1A is a block diagram illustrating a road map creation system according to an embodiment. FIG. 1B is a block diagram illustrating a server device and a data processing device of the road map creation system according to the embodiment. FIG. 2 is a diagram showing transmission data sent from the in-vehicle device of the road map creation system of the embodiment to the server device. FIG. 3 is a data chart showing a modification of transmission data sent from the in-vehicle device to the server device of the road map creation system of the embodiment. FIG. 4 is a diagram illustrating a vehicle database of the road map creation system according to the embodiment. FIG. 5 is a flowchart showing the operation of the data processing apparatus of the road map creation system of the embodiment. FIG. 6 is a diagram showing a photographed image of the road map creation system of the embodiment. FIG. 7 is a diagram for explaining the process of expanding the road map coordinates according to the embodiment of the road map creation system. FIG. 8 is a diagram for explaining the feature extraction process of the road map creation system according to the embodiment. FIG. 9A is a diagram illustrating a process of copying captured image data to a high-definition data series in the road map creation system of the embodiment. FIG. 9B is a diagram illustrating a process of copying photographed image data to a high-definition data series in the road map creation system according to the embodiment. FIG. 9C is a diagram illustrating a process of copying captured image data to a high-definition data series in the road map creation system according to the embodiment. FIG. 10 is a diagram for explaining the principle of generating high-definition image data from a plurality of image data in the road map creation system of the embodiment.

In the above-described conventional technology, it takes a lot of time and cost to create road map data for various places in the country using tens or hundreds of road measurement vehicles. Furthermore, since road construction is frequently performed at various locations, there is a limit to using a fixed number of road measurement vehicles to always prepare the latest road map data.

In addition, at present, the vehicle is often equipped with a camera for photographing the surroundings of the vehicle, such as a drive recorder or a parking guide camera. Furthermore, the vehicle often has a function of communicating with a server device such as a navigation device. Therefore, it is also possible to consider collecting captured images from a plurality of vehicles and generating map data as in the prior art disclosed in

Patent Documents

1 and 2. However, the captured image of the vehicle has a low resolution and cannot be used for creating high-definition road map data as it is.

Hereinafter, the road map creation system of the embodiment will be described in detail with reference to the drawings.

FIG. 1A is a block diagram showing a road map creation system according to an embodiment of the present invention.

The road map creation system of the present embodiment includes a plurality of in-vehicle devices 10 respectively mounted on a plurality of vehicles 100, a server device 30 that is communicably connected to the in-vehicle devices 10 via a network, and road map data. And a data processing device 50 to be created.

FIG. 1B is a block diagram showing the server device 30 and the data processing device 50. The server device 30 and the data processing device 50 may be configured by a single computer, or may be configured to perform processing distributed by more computers.

The in-vehicle device 10 includes a camera 11 that is one or a plurality of photographing devices, a position acquisition unit 12, a communication unit 13, a display 14, a control unit 15, an interface (I / F) 16, and a vehicle speed sensor 17. And. The in-vehicle device 10 can be configured by adding a control function to an existing device such as a car navigation device.

The in-vehicle device 10 does not include the camera 11, the position acquisition unit 12, and the vehicle speed sensor 17, and may be configured to input data from the camera 11, the position acquisition unit 12, and the vehicle speed sensor 17 mounted on the vehicle 100. Good. Moreover, the vehicle-mounted apparatus 10 is good also as a structure by which the display 14 was abbreviate | omitted.

The camera 11 is mounted so as to photograph the outside of the passenger compartment such as the front or rear of the vehicle 100. The camera 11 performs a photographing process outside the passenger compartment in accordance with a command from the control unit 15, and sends captured image data to the control unit 15.

The position acquisition unit 12 calculates the current position using, for example, a GPS (Global Positioning System), an accelerator sensor, and a vehicle speed sensor. The acquired current position data is sent to the control unit 15.

The vehicle speed sensor 17 is, for example, a wheel speed sensor, and sends current vehicle speed data to the control unit 15.

The communication unit 13 can transmit data to the server device 30 via the network. The communication unit 13 may be configured to perform data transmission via a mobile communication network and the Internet.

The interface 16 connects the control unit 15 and each unit so that data can be input / output. The interface 16 functions as an input unit that inputs a captured image.

The control unit 15 comprehensively controls each unit of the in-vehicle device 10. When the captured image is input from the camera 11, the control unit 15 inputs position information indicating the position of the vehicle 100, that is, the in-vehicle device 10 and vehicle speed information indicating the speed of the vehicle 100 from the position acquisition unit 12 and the vehicle speed sensor 17. And the vehicle ID (identification signal) are associated with each other and transmitted to the server device 30 via the communication unit 13.

FIG. 2 is a data chart showing transmission data sent from the in-vehicle device to the server device. FIG. 3 is a data chart showing a modification of the transmission data of the in-vehicle device.

The transmission data 20 of the in-vehicle device 10 includes data of each item in FIG. That is, the captured image data, captured image acquisition date and time, position information and vehicle speed information, and data format information are included. The data serial number is a serial number assigned to each vehicle 100. The vehicle ID and camera number are information for identifying the vehicle 100 and information for identifying the camera 11 when a plurality of cameras 11 are mounted on the vehicle 100. The vehicle ID and camera number are registered in the vehicle database 33 (see FIG. 4). The vehicle ID and the camera number are identification information that indicates the mounting position of the in-vehicle camera on the vehicle 100. The transmission data 20 may include information on the traveling direction of the vehicle.

As shown in FIG. 3, the transmission data 20A sent from the in-vehicle device 10 to the server device 30 includes camera position information (information such as front and rear, left and right, height, and elevation angle) instead of the camera number. Also good.

The server device 30 receives and collects data of captured images including roads from the plurality of in-vehicle devices 10. The server device 30 includes a communication unit 31, a control unit 32, a vehicle database 33, and a captured image database 34.

The communication unit 31 receives transmission data of the in-vehicle device 10 including a captured image via the network.

The photographed image database 34 has a large-scale storage device and stores photographed images sent from a plurality of in-vehicle devices 10. The captured image is stored in association with additional information. The additional information includes shooting position (latitude and longitude) information, camera position information, and shooting date and time information. The additional information may include vehicle speed information at the time of shooting, an image data format, a serial number, and the like. The vehicle speed information can be used to supplement the shooting position information.

The control unit 32 registers a photographed image including a road in the photographed image database 34 in association with necessary information from the information in the vehicle database 33 and the transmission data of the in-vehicle device 10.

FIG. 4 is a data chart showing the vehicle database.

In the vehicle database 33, information on a plurality of vehicles 100 on which the in-vehicle device 10 is mounted is registered. The information on the vehicle 100 includes a vehicle ID (identification information) for identifying each vehicle 100, the number of cameras 11 mounted so that a road can be photographed, and a mounting position of each camera 11 mounted on the vehicle 100. Camera position information to be displayed.

The camera position information includes information indicating the horizontal mounting position of the vehicle 100, information on the height position of the camera 11, information on the elevation angle of the camera 11, and the like.

The control unit 32 of the server device 30 reads the camera position information corresponding to the vehicle ID and the camera number from the vehicle database 33 when receiving the transmission data 20 from the in-vehicle device 10. Then, the read camera position information and captured image data are registered in the captured image database 34 in association with each other.

As shown in FIG. 1B, the data processing device 50 includes an image search unit 51, a feature extraction unit 52, a high-definition processing unit 53, a position designation unit 54, a map data management unit 55, and a road map database 56. Yes.

The position designation unit 54 designates position information indicating the position of a point where a high-definition road image is created or a point where a high-definition road image is updated. The position designation unit 54 may designate the position information based on input information from the outside. For example, a configuration may be adopted in which an operator inputs a command to create road map information for a district from the outside, and the position designation unit 54 designates position information based on the command. Alternatively, the position specifying unit 54 may be configured to specify position information using a predetermined algorithm. For example, the position specifying unit 54 may be configured to sequentially specify position information at predetermined meter intervals in a predetermined area on the earth specified from the outside. In addition, the position specifying unit 54 may be configured to specify a point that is reflected in the captured image received by the server device 30 from the in-vehicle device 10.

The image search unit 51 searches the photographed image database 34 to find a photographed image including the image of the spot designated by the position designation unit 54. The image search unit 51 can find the corresponding captured image from the position information associated with the captured image. Further, the image search unit 51 may find a corresponding captured image in consideration of information on the mounting position of the camera 11 and information on the traveling direction of the vehicle. For example, if the camera 11 captures the front, the position of the subject area of the camera 11 can be calculated by adding the amount of displacement in the traveling direction to the position information.

The feature extraction unit 52 extracts a feature contained in the image in order to detect the positional deviation of the plurality of captured images with high accuracy. The characteristic object is a fixed object fixed on the land, and for example, an artificial or natural mark on the road can be applied. Marks on roads that are artificially included include road markings (white lines, orange lines, etc.), cracks generated in road markings, manholes, and the like. A natural mark on a road includes a crack or depression in the road. In addition, as a feature, a traffic facility such as a guardrail or a sign, or a building facility such as a building or a signboard can be applied. The feature extraction unit 52 extracts, for example, an edge of a feature object.

The high-definition processing unit 53 creates a high-definition road image more than each captured image by using a plurality of captured images with slightly different shooting targets. A method for creating a high-definition image will be described later.

The map data management unit 55 registers the high-definition road image data of each point created by the high-definition processing unit 53 in the road map database 56 together with the position information. Thereby, high-definition road images of a plurality of sections are collected, and road map data is generated.

The road map database 56 has a large-scale storage device, and stores position information and high-definition images of roads in association with each other.

<Data processing>
Subsequently, processing contents of the data processing device 50 will be described.

FIG. 5 is a flowchart showing an example of the operation of the data processing apparatus.

In step S1, the position designation unit 54 designates a point where a high-definition road image is created. The point designation method is as described in the explanation part of the position designation unit 54.

In step S2, the image search unit 51 searches the photographed image database 34 for a plurality of photographed images (hereinafter, also referred to as “common photographed images”) including images of common points. The search method is as described in the explanation part of the image search unit 51.

In step S3, the feature extraction unit 52 first reads the photographed image selected in step S2 and additional information (position information, information on the in-vehicle position of the camera, etc.), expands the photographed image into road surface coordinates, Further, the luminance is normalized. This process will be described with reference to FIGS.

FIG. 6 is a diagram illustrating an example of a captured image. FIG. 7 is a diagram for explaining the development process to road surface coordinates.

As shown in FIG. 6, in the photographed image, the vicinity of the road is displayed large, and the distance from the road is displayed small. Of the additional information associated with the photographed image, the height and angle of the photographing position from the road can be specified by the information on the camera position. From these pieces of information, it is possible to obtain a mapping function from a captured image to an image when viewed in plan when the imaging target is considered as a plane.

The feature extraction unit 52 obtains the above mapping function, further recognizes the road image from the captured image, and develops the road image into planar road surface coordinates as shown in FIG. The developed image has different resolutions on the near side and the far side, and the resolution on the far side is low. FIG. 7 shows the size of the pixel 81 in the range A and the size of the pixel 82 in the range B.

As an example, the normalization of the brightness is by calculating the average value of the brightness of all the pixels on the road and subtracting or adding the brightness of all the pixels uniformly so that the average value becomes a predetermined reference value. Done. Thereby, the brightness | luminance of the image image | photographed when it was dark and the image image | photographed when it was bright can be match | combined with standard brightness | luminance. In addition, luminance normalization may be performed by adding correction according to the perspective of the road. The feature extraction unit 52 performs such normalization processing. In addition, the feature extraction unit 52 may perform a normalization process on the hue or saturation according to the characteristics of the camera 11.

In step S4, the feature extraction unit 52 extracts a feature in the image from the developed image. The feature extraction process will be described with reference to FIG.

FIG. 8 is a diagram for explaining an example of feature extraction processing.

The feature extraction unit 52 performs, for example, edge detection in an image to extract feature objects. Here, the extraction of the feature object may be performed on a mark on the road and an area on the near side (area with high resolution) in the captured image. For example, when extracting a white line on a road, the feature extraction unit 52 performs finer extraction on the near white line f1 (see FIG. 7) in the captured image. The feature extraction unit 52 also extracts a crack or blur of the white line f1. By comparing the feature extracted in this way, it is possible to determine whether or not the white line f1 is the same among a plurality of captured images.

The feature extraction unit 52 performs development on road surface coordinates and feature extraction for each common captured image found in step S2. The feature object extraction data and the image data developed to the road surface coordinates are sent to the high-definition processing unit 53.

It should be noted that the search for the common captured image (step S2), the development from the captured image to the road surface coordinates (the first stage of step S3), and the feature extraction (step S4) need not be performed in this order. For example, feature extraction may be performed sequentially on a plurality of received captured images. The extracted feature data may be stored in association with the photographed image. The extracted feature data can also be used when searching for a plurality of common captured images.

In step S5, the high-definition processing unit 53 aligns a plurality of common captured images based on the extracted features. The common captured image to be aligned is an image developed on road surface coordinates. The alignment is performed so that the same feature overlaps between a plurality of common captured images. The position information using GPS includes a relatively large error. For this reason, the alignment accuracy between a plurality of common captured images is increased by performing the alignment so that the feature objects in the image overlap each other.

In step S5, if there is a common captured image in which a common feature is not found, this image may be excluded and processed.

In step S6, the high-definition processing unit 53 copies the pixel data of a plurality of common captured images to the high-definition data series, and weights and adds the pixel data of each common captured image. These processes will be described with reference to FIGS. 9A to 9C and FIG.

FIG. 9A to FIG. 9C are diagrams for explaining processing for copying photographed image data to a high-definition data series. FIG. 9A shows an enlarged display of a part of pixels 83 of one common photographed image developed on road surface coordinates. FIG. 9B shows an enlarged display of a part of the arrangement of high-definition pixels corresponding to the high-definition data series. FIG. 9C shows an example of the pixel data (a) copied to a high-definition data series.

The common captured image developed on the road surface coordinates includes a large pixel 83 as shown in FIG. 9A. On the other hand, the data series of the high-definition image is a data series corresponding to the arrangement of the fine pixels 90 as shown in FIG. 9B. In FIG. 9B and FIG. 9C, only one of the plurality of pixels 90 is denoted by a symbol in order to avoid complexity.

In step S6, first, as shown in FIGS. 9A to 9C, the fine pixels 90 of the high-definition image and the large pixels 83 of the common captured image are superposed. Then, the pixel data of the pixel 83 where the common photographed image is superimposed is copied as fine data of each pixel 90. Thereby, the data of one common photographed image is copied into a high-definition data series.

Note that, depending on the position of the pixel 90 of the high-definition image, a plurality of adjacent pixels 83 may overlap with one pixel 90. In this case, pixel data of the pixel 83 having a larger overlapping range among the plurality of pixels 83 may be copied as data of one pixel 90. Alternatively, the pixel data of a plurality of pixels 83 may be averaged and copied as data of one pixel 90.

Further, in step S6, the pixel data of a plurality of common captured images are weighted and added as the data of the pixels 90 of the high-definition image. The weighting is preferably performed so that the weighting value increases according to the reliability of the pixel data. For example, when adding pixel data of n common captured images, the reliability of the pixel data of each common captured pixel is assumed to be the same, and the data value of each pixel is weighted by 1 / n. Good. When the pixel data of one new common photographed image is already added to the pixel data obtained using (m−1) common photographed images, the former is heavily weighted (for example, (m− 1) / m), and the latter may be given a small weight (for example, 1 / m).

FIG. 10 is a diagram for explaining the principle of generating high-definition image data from a plurality of image data. FIG. 10 shows a pixel 87 (4 × 4 pixels) in a section of the first image, a pixel 88 (4 × 4 pixels) in a section of the second image, a plurality of pixels 87, and a plurality of pixels. 88 is superimposed. The pixel 87 of the first image and the pixel 88 of the second image are pixels obtained by imaging the same subject with a slight shift. In FIG. 10, the positional deviation is shown as a deviation from the origin O on the xy plane. In order to avoid complications, only one of each of the plurality of pixels 87 and 88 is given a reference numeral.

One pixel data is a value obtained by averaging the luminance and color of the object area corresponding to this pixel. For this reason, when there are the first image pixel 87 and the second image pixel 88 obtained by photographing the same subject area slightly shifted, the two positions are aligned and the two are overlapped. It is possible to estimate the brightness and color of a region finer than each of the pixels 87 and 88. For example, as shown in FIG. 10C, a range where one pixel 87 of the first image and one pixel 88 of the second image overlap is smaller than one pixel 87 or pixel 88. The actual luminance and color information of the subject in this overlapped range is reflected in the data of the pixels 87 and 88 that overlap this region. Therefore, the luminance and color of the area finer than the pixels 87 and 88 are estimated by weighting and adding the luminance and color of the pixels 87 and 88 slightly shifted in position. The more images that are taken with respect to the same subject, the more the images are taken, the more accurately the brightness and color of the fine area are estimated, and a high-definition image can be obtained.

In step S6, based on such a principle, the pixel data of a plurality of common captured images with low resolution are aligned and weighted to add the luminance and color of a region smaller than the pixels of the common captured image. By estimation, a high-definition road image can be obtained.

In step S7, the map data management unit 55 registers the high-definition road image obtained in step S6 together with the position information in the road map database. Thus, road map data in which high-definition road images are continuous is generated.

The data processing device 50 can create road map data covering many districts by performing the processing of FIG.

As described above, according to the road map creation system of the present embodiment, for example, a large number of captured images are collected using an in-vehicle device mounted on a general automobile, and high-definition road map data is obtained from these. Can be created. Therefore, it is not necessary to prepare many special road measurement vehicles, and road map data can be created efficiently.

Note that the present invention is not limited to the specific contents described in the above embodiment. For example, in the photographed image search processing, it has been described that a photographed image including an image of the same point is searched from position information and information of a photographing direction. However, a photographed image that is estimated to include an image of approximately the same point from only position information. You may make it look for. In this case, in the common photographed image that has been found, the ratio of the photographed image that does not include the image of the same point is high, but by the feature extraction process and the alignment process based on the feature, The captured image can be excluded and processed.

In the above-described embodiment, an example of processing for creating a high-definition road image from a plurality of captured images is shown. However, the processing content of the embodiment is merely an example, and the image was captured with a slight shift in position. Various known image techniques for creating a high-definition image using a plurality of images may be applied.

In the above embodiment, it has been explained that road map data is finally created. However, the road map data does not need to include road images of all sections, and road images in which some sections are missing. It may be road map data.

Further, although cases have been described with the above embodiment as examples where the present invention is configured by hardware, the present invention can also be realized by software in cooperation with hardware.

The present invention can be used in a system for creating a road map.

DESCRIPTION OF SYMBOLS 10 In-vehicle apparatus 11 Camera 12 Position acquisition part 13 Communication part 14 Display 15 Control part 16 Interface 20, 20A Transmission data 30 Server apparatus 31 Communication part 32 Control part 33 Vehicle database 34 Shooting image database 50 Data processing apparatus 51 Image search part 52 Features Extraction unit 53 High-definition processing unit 54 Position specifying unit 55 Map data management unit 56 Road map database

Claims

A plurality of in-vehicle devices that are respectively mounted on a plurality of vehicles and transmit a plurality of captured images of the plurality of vehicles to a server device;
A data processing device for creating a road image using the plurality of captured images transmitted;
With
The data processing device includes:
An image search unit for searching for a plurality of common photographed images including images of common points from the plurality of photographed images transmitted;
A feature extraction unit for extracting features in the plurality of transmitted captured images;
Based on the feature, the positions of the plurality of common photographed images are aligned, and from each pixel information of the plurality of common photographed images that have been aligned, a road image with higher definition than each of the plurality of common photographed images is obtained. A high-definition processing unit to be generated;
Road map creation system with
The plurality of photographed images are respectively photographed by a plurality of photographing devices respectively mounted at a plurality of mounting positions of the plurality of vehicles.
The plurality of in-vehicle devices include a plurality of pieces of position information respectively indicating positions of the in-vehicle devices when the plurality of photographed images are photographed, a plurality of pieces of identification information indicating the plurality of mounting positions, and the plurality of photographed images. To the server device,
The road map creation system according to claim 1, wherein the image retrieval unit retrieves the plurality of common captured images using the plurality of position information and the plurality of mounting position information.
The plurality of captured images include a road,
The road map creation system according to claim 1, wherein the feature extraction unit extracts a mark on the road as the feature.
The road map creation system according to claim 1, wherein the data processing device further includes a map data management unit that collects the high-definition road images of a plurality of sections to create high-definition road map data.
A data processing apparatus that creates a road image using a plurality of captured images captured by a plurality of vehicles and collected by a server device,
An image search unit for searching for a plurality of common photographed images including an image of a common spot from the collected photographed images;
A feature extraction unit for extracting features in the collected captured images;
Based on the feature, the positions of the plurality of common photographed images are aligned, and from each pixel information of the plurality of common photographed images that have been aligned, a road image with higher definition than each of the plurality of common photographed images is obtained. A high-definition processing unit to be generated;
A data processing apparatus comprising:
An image search unit for searching for a plurality of common photographed images including a common point image from a plurality of photographed images collected in the server device;
A feature extraction unit for extracting features in the collected captured images;
A high-definition image that generates a high-definition road image that is higher than each common-captured image from the information of each pixel of the plurality of common-captured images that are aligned with each other based on the features. A data processing device having a processing unit, providing one captured image of the plurality of captured images, and being mounted on a vehicle,
An input unit for inputting a photographed image outside the vehicle from a photographing device mounted on the vehicle;
A communication unit that transmits the outdoor photographed image as the one photographed image to the server device;
In-vehicle device equipped with.
A control unit that causes the photographing apparatus to photograph a region including a road;
The in-vehicle device according to claim 6, wherein the communication unit transmits a captured image including the road as the one captured image to the server device.
A position acquisition unit that acquires position information indicating the position of the in-vehicle device;
The communication unit transmits data in which the position information of the vehicle when the outdoor captured image is captured and identification information that identifies the mounting position of the imaging device on the vehicle are transmitted to the outdoor captured image. The in-vehicle device according to claim 6.