Information processing apparatus and information processing method

Abstract

To provide an information processor capable of properly handling of occluded areas on roads, and an information processing method.SOLUTION: The information processor includes: a first information conversion unit 15 that estimates a first road area from images taken of the road; a second information conversion unit 17 that estimates a second road area from map information stored in a map information storage unit 11; and a difference extraction unit 16 that extracts difference information representing a difference between the first road area and the second road area. In the information processing method, a computer estimates the first road area from the images taken of the road and estimates the second road area from the map information to extract the difference information representing a difference between the first road area and the second road area.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an information processing apparatus and an information processing method. For example, it relates to an information processing apparatus and an information processing method for handling map data for autonomous driving and advanced safety driving systems.

Background Art

[0002] For autonomous driving and driving support, there is a known method of constructing a highly accurate map by collecting road information with a vehicle for map generation or manually, for a map that holds road and lane information in a network form. In this method, since a considerable amount of human labor is generated, only information on some roads such as highways can be constructed, the comprehensiveness of the constructed map information is reduced, and there is a problem that the range in which the map can be provided for autonomous driving and driving support becomes narrow.

[0003] In response to this problem, for example, Patent Document 1 discloses a method of updating a map using data collected from an imaging device mounted on a general vehicle. Patent Document 2 discloses a technique for improving the accuracy of a map using aerial photographs and navigation map data. Further, Patent Document 3 discloses a method of correcting a map using aerial photograph data and substrate road map data.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the conventional technology has a problem that it cannot appropriately handle a shielding area where a road is shielded and does not appear in an image.

[0006] In aerial photographs, satellite images, and videos captured by imaging devices mounted on general vehicles, there are occluded areas where the road is blocked by tunnels, overpasses, high-rise building groups, etc. Therefore, there is a problem that incorrect information is reflected when the road information extracted from the captured image is reflected on the map.

[0007] The present invention has been made to solve such problems, and an object thereof is to provide an information processing apparatus and an information processing method capable of appropriately handling occluded areas of roads.

Means for Solving the Problems

[0008] An example of the information processing apparatus according to the present invention is a first information conversion unit that estimates a first road area from an image of a road, a second information conversion unit that estimates a second road area from map information stored in a map information storage unit, a difference extraction unit that extracts difference information representing a difference between the first road area and the second road area, and includes.

[0009] An example of the information processing method according to the present invention is a computer estimates a first road area from an image of a road, estimates a second road area from map information, and extracts difference information representing a difference between the first road area and the second road area.

Effects of the Invention

[0010] The information processing apparatus and the information processing method according to the present invention can appropriately handle occluded areas of roads.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Mode for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present invention will be described based on the accompanying drawings. [Example 1] FIG. 1 shows the configuration of the information processing apparatus according to Example 1. The information processing apparatus executes the information processing method described in this specification.

[0013] The image acquisition unit 12 acquires a satellite image or an aerial image belonging to the range specified by the latitude / longitude specifying unit 13 from the image information storage unit 10 that stores satellite images and aerial photographs. Here, the satellite image is, for example, an image obtained by photographing the ground from a camera mounted on a satellite, and restored so as to overlook the ground data. The aerial image is, for example, an image obtained by photographing the ground from a camera mounted on an aircraft, and restored so as to overlook the ground data.

[0014] The latitude / longitude specifying unit 13 may specify the latitude and longitude so that the image to be acquired is extracted, or may substitute the latitude and longitude with an identifier or the like that substitutes for the latitude and longitude.

[0015] In the first information conversion unit 15, a road area (first road area) is estimated from an image obtained by imaging a road. For example, the satellite image or aerial image obtained by the image acquisition unit 12 is processed to estimate the road area. The first road area may be represented as an image.

[0016] The road area is estimated, for example, as an attribute assigned to each pixel of the image, and includes information indicating whether each pixel is a road. This information can be represented as an image. Alternatively, the road area may be extracted in a network form represented by nodes and links. In that case, the extracted information may be represented by a set of pixels.

[0017] In the map information acquisition unit 14, map information (for example, information in a network format) of the area specified by the latitude / longitude specifying unit 13 is acquired from the map information storage unit 11.

[0018] The second information conversion unit 17 estimates a road area (second road area) from the map information stored in the map information storage unit 11. For example, it outputs data in the same format as that output by the first information conversion unit 15. That is, when the first information conversion unit 15 outputs an image, the second information conversion unit 17 also generates an image showing the road area. When the first information conversion unit 15 outputs data in a network format, the second information conversion unit 17 also outputs data in a network format.

[0019] The difference extraction unit 16 extracts difference information representing the difference between the above two road areas (first road area and second road area).

[0020] The structure information verification unit 18 verifies the structure related to the road based on the extracted difference information and outputs the result. Hereinafter, unless otherwise specified, it is assumed that the formats output by the first information conversion unit 15 and the second information conversion unit 17 are images for explanation, but the outputs of both may be in a network form. The difference extraction unit 16 may compare images with each other or compare networks with each other.

[0021] The information processing apparatus in FIG. 1 can be configured using, for example, a known computer. The information processing apparatus includes, for example, arithmetic means and storage means. The arithmetic means includes, for example, a processor, and the storage means includes storage media such as a semiconductor memory device and a magnetic disk device.

[0022] Also, the computer may include input / output means. The input / output means includes, for example, input devices such as a keyboard and a mouse, output devices such as a display and a printer, and communication devices such as a network interface.

[0023] The storage means may store a program. By the processor executing this program, the computer may execute the functions described in this embodiment. That is, the computer functions as each component shown in FIG. 1, thereby operating as an information processing apparatus, that is, executing an information processing method.

[0024] The configurations of the first information conversion unit 15 and the second information conversion unit 17 are shown in FIG. 2. In the first information conversion unit 15, using the image acquired by the image acquisition unit 12 as an input, it is estimated whether, for example, each pixel of the image is a road area or not. Here, for example, a pre-trained model 150 such as a neural network pre-trained by machine learning can be used for estimation by techniques such as semantic segmentation.

[0025] In the following description, the term "road" is a general term for a road consisting of a single lane and a road consisting of multiple lanes. For example, for a road consisting of multiple lanes, it may be processed as a single road as a whole, or may be processed as multiple parallel roads.

[0026] The estimation of the road area is performed by the first road area estimation unit 151. Additionally, there may be a first road boundary estimation unit 152 that estimates the boundary between roads or between a road and a non-road area, or a first road indication area estimation unit 153 that estimates white lines, yellow lines, zebra zones, etc. In order to estimate more detailed information, there may be a first lane boundary estimation unit 154 and / or a first lane area estimation unit 155 that divide the road area into lanes and estimate it together with the lane boundaries.

[0027] The map information storage unit 11 stores, for example, road data included in map data for navigation. The data here has the shape of the road stored in a network form. For example, its elements are stored in a state representing nodes, links, the shape of those links, and the attributes of the road. For example, the ends of the road are represented as nodes, and the road is represented as a link connecting two nodes. Information indicating the attributes of the road (such as the width of the road or lane) may be associated with the link.

[0028] Similar to the first information conversion unit 15, the second information conversion unit 17 may have, in addition to the second road area estimation unit 171, a second road boundary estimation unit 172 that estimates the boundary between a road and a non-road area or between a road and another road area. In order to estimate more detailed information, there may be a second lane boundary estimation unit 174 and / or a second lane area estimation unit 175 that divide the road area into lanes and estimate it together with the lane boundaries. If the map holds more detailed information such as yellow line and white line information, there may be a second road indication area estimation unit 173.

[0029] For example, when the given network information indicates the position of the center line of a road, the second road area estimation unit 171 can first convert the network information into image-formatted information, and then estimate the road area by expanding the pixels on both sides of the center line. The expansion can be performed, for example, by constructing a disk with a radius equal to half the number of pixels corresponding to the lane width for each pixel constituting the center line, while considering the number of lanes, and overlapping the disks. The actual number of pixels to be expanded can be determined according to, for example, the distance per pixel of the image obtained by the first information conversion unit 15. When two parallel roads overlap after expansion, the center line of the overlapping area may be treated as the road boundary. The processing for the case where there is a vertical relationship in altitude on the road (for example, a flyover) can be appropriately designed by those skilled in the art. For example, the road with a higher altitude may be preferentially expanded.

[0030] The function of the difference extraction unit 16 is shown in FIG. 4. In this embodiment, among the estimation units shown in FIG. 2, the results of the first road area estimation unit 151 and the second road area estimation unit 171 are used. However, the processing results of the first road boundary estimation unit 152 and the second road boundary estimation unit 172, the first road indication area estimation unit 153 and the second road indication area estimation unit 173, the first lane boundary estimation unit 154 and the second lane boundary estimation unit 174, etc. may also be used as information to improve the accuracy of difference extraction.

[0031] The difference extraction unit 16 extracts the difference between the images obtained from the first information conversion unit 15 and the second information conversion unit 17. The images obtained from the two information conversion units may have a deviation in the positions of the roads due to the distortion of the original satellite images or aerial photographs, or the deviation in the positions of the networks of the maps registered in the map information storage unit 11. Therefore, first, the rough alignment of the positions of the roads is performed by the initial position correction unit 161. The specific processing for the alignment between the two images can be appropriately designed based on known techniques.

[0032] To perform this alignment, the second information conversion unit 17 may refer to the original network information for adjustment and regenerate the image. Next, according to the comparison format (162), when comparing based on the road, in the road presence / absence determination unit 163 and the road width comparison unit 165, and when comparing based on the lane, in the lane presence / absence determination unit 164 and the lane width comparison unit 166, the presence / absence of the road on the image, the comparison between road widths, or the presence / absence of the lane and the comparison of lane widths are performed.

[0033] In the comparison of the presence / absence of the lane and the comparison of the lane width, the process of recognizing the road based on the first road region and the second road region may be performed. Also, after recognizing the road, the process of recognizing the endpoints and width of the road may be performed. Furthermore, the process of determining the correspondence relationship of the road between each road region may be performed. These specific processes can be appropriately designed based on known techniques. As a result, the presence / absence of the road or the width of the road can be compared.

[0034] The road presence / absence determination unit 163 determines the difference regarding the presence / absence of the road in the first road region and the second road region (both are images in the following example). For example, when there is road information in one image and there is no corresponding road information in the other image, it is determined that there is a difference regarding the presence / absence of the road.

[0035] When there is road information associated with both images, the road width comparison unit 165 compares the widths of the roads on the image, obtains the difference in width, and outputs it. Regarding the output of the difference, it may be processed after considering conversion errors and the like in the data input of both sides, and the output unit may be the number of pixels or the number of lanes corresponding thereto.

[0036] The lane presence / absence determination unit 164 determines the difference regarding the presence / absence of the lane in the first road region and the second road region. When there is lane information associated with both images, the lane width comparison unit 166 compares the widths of the lanes on the image, obtains the difference in width, and outputs it.

[0037] The difference extraction unit 16 summarizes this information as difference information and transmits it to the structure information verification unit 18. As described above, the difference information includes difference presence / absence information indicating the presence or absence of a difference and difference amount information indicating the amount of the difference. Here, the "presence or absence of a difference" means, for example, whether there is a difference regarding the presence or absence of a road and whether there is a difference regarding the width of the road, and the "amount of the difference" means, for example, the magnitude of the difference regarding the difference in the width of the road.

[0038] Examples of the unit that outputs the difference include a part of a road or a lane, a set of pixels on an image, a node or a link in a network, etc. Hereinafter, the unit that outputs the difference is referred to as a section. A section is, for example, an image divided according to a predetermined standard (for example, into equal-sized squares). Also, it is assumed that which part of the information in the original image information storage unit 10 and the map information storage unit 11 the output section corresponds to can be determined by tracing the conversion process so far.

[0039] The configuration of the structure information verification unit 18 is shown in FIG. 5. In this example, the structure information verification unit 18 includes a map information verification unit 20.

[0040] The map information verification unit 20 classifies and outputs the type of difference based on the comparison result of the difference extraction unit 16 and the structure information of the road or lane. Note that hereinafter, the function will be described focusing on the road, but the road may be replaced with a lane. The structure information verification unit 18 checks with the map information verification unit 20 whether or not occlusion has occurred in the target section.

[0041] The map information verification unit 20 refers to the information in the corresponding map information storage unit 11 in the section where a difference has occurred and determines whether or not there is an occlusion area (that is, whether or not the road is occluded by other structures). For example, as illustrated in FIG. 11, if the road is inside a tunnel, that part is determined to be an occlusion area. Also, for example, if the road is an access road to a junction or an expressway, that part is determined to be an occlusion area. In the part where the roads intersect three-dimensionally, the occlusion area may be determined considering the widths of both roads.

[0042] Here, if there is occlusion as the output, 1 will be output, and if not, 0 will be output, and it will be output as the variable "#" in FIG. 5. However, the value of the variable # may be provided with levels step by step depending on the degree of occlusion or the like.

[0043] First, when it is determined in the difference extraction unit 16 that there is a difference (S1: Yes), if there is a road on the image given from the first information conversion unit 15 and there is no road on the image given from the second information conversion unit 17 (S2), signal A is output as the classification result.

[0044] When it is determined that there is a difference (S1: Yes), and there is no road on the image given from the first information conversion unit 15 and there is a road on the image given from the second information conversion unit 17 (S3), signal B is output. In particular, in the map information verification unit 20, if there is no occlusion for the road related to the difference, a signal of B0 is output, and if there is occlusion, a signal of B1 is output.

[0045] When it is determined that there is a difference (S1: Yes), and there is a road on the image given from the first information conversion unit 15 and there is also a road on the image given from the second information conversion unit 17 (S4), signal C or D is output. That is, it is a case where although there are roads in both, it is estimated that the number of lanes or the road width registered for the road is different.

[0046] In this case, if the width (or the number of lanes) of the road observed on the image given from the first information conversion unit 15 is larger than that given from the second information conversion unit 17 (S5: Yes), in the map information verification unit 20, signal D is output, and if not (S5: No), signal C is output.

[0047] In particular, for signal C, in the map information verification unit 20, if there is no occlusion for the road related to the difference, a signal of C0 is output, and if there is occlusion, a signal of C1 is output. Similarly, for signal D, in the map information verification unit 20, if there is no occlusion for the road related to the difference, a signal of D0 is output, and if there is occlusion, a signal of D1 is output.

[0048] If the difference extraction unit 16 determines that there is no difference (S1: No), and if there are roads on both sides (S6), then signal E is output. In particular, if there is no occlusion, signal E0 is output, and if there is occlusion, signal E1 is output.

[0049] If the difference extraction unit 16 determines that there is no difference (S1: No), and if there are no roads on both sides (S7), then signal F is output.

[0050] In this way, the structure information verification unit 18 distinguishes and outputs the type of difference by checking the occlusion status in the map of the section where the difference occurred.

[0051] The information processing device may include a difference reliability output unit 30 as shown in FIG. 10, and the difference reliability output unit 30 may determine the reliability of the difference according to the result of the process in FIG. 5.

[0052] FIG. 6 shows the relationship between the difference information and the reliability. The difference reliability output unit 30 outputs the reliability for each of the difference presence / absence information and the difference amount information. In this embodiment, the difference reliability output unit 30 determines the reliability based on the first road region, the second road region, and the presence / absence of occlusion. With such a configuration, the reliability is determined while appropriately considering the presence / absence of occlusion.

[0053] The reason why only the reliability of A is "medium" is that the data to be referred to is not in the map information storage unit 11.

[0054] The data of "―" indicates that there is no difference, that is, there is no data to be updated. The reliability in the case of having occlusion (B1, C1, D1, E1) is lower than that in the case of having no occlusion (B0, C0, D0, E0). This is because road recognition from the image may be inaccurate due to occlusion.

[0055] Regarding road recognition from an image (processing based on the first road area), when there is occlusion, a part of the road may not appear in the image, so the recognized road is likely to be narrower than the actual road and less likely to be wider than the actual road. Therefore, the confidence level in the case of C1 is low, and the confidence level in the case of D1 is high (higher than C1).

[0056] The information processing apparatus may include a map information update unit 31 as shown in FIG. 7. The map information update unit 31 updates the map information in the map information storage unit 11 based on the difference information (for example, according to the results of the structure information verification unit 18 and the difference confidence level output unit 30). More specifically, it determines whether to update the map information based on the confidence level of the difference presence / absence information and the confidence level of the difference amount information. And when it is determined to update, the map information is updated. According to such a configuration, the map information can be appropriately updated.

[0057] In this embodiment, the map information is updated according to the results classified in FIGS. 5 and 6.

[0058] Note that a reference example of the update is shown below. It is assumed that the data in the image information storage unit 10 to be compared is newer than the data in the map information storage unit 11. When the data in the image information storage unit 10 is older, the map information update unit 31 does not update the map information. However, as a modification, the update may be performed even when the data in the image information storage unit 10 is older. Also, as another modification, when the newness of the data is approximate (for example, the time difference is within a predetermined range) or the confidence level of the map information is low, the map information may be updated. Note that this "confidence level of the map information" may be determined by the processing described later or by using other appropriate criteria.

[0059] First, an example of additionally registering whether the reliability of the presence or absence of information registered on the map is high or not will be shown. In FIGS. 5 and 6, the map information update unit 31 may determine to update the map information in the case of A (that is, when there is a road in the first road area and there is no corresponding road in the second road area). As a specific example, information about the road may be added to the map information. By doing so, when a new road is discovered from an image, the information can be reflected in the map information. As a modification, in the case of A, it may be determined not to update the map information.

[0060] For cases B0, C0, D0, and D1 where it is considered highly likely that the presence or absence of the extracted difference is correct, information indicating that the reliability of the map information is low is given to the map. On the other hand, for B1, C1, and E1 where it is considered highly likely that the extracted difference is incorrect, the reliability of the map information is not updated, or information indicating that the reliability of the map information is high as a result of verification with the image information storage unit 10 is given.

[0061] That is, in the case (B) where there is a road in the second road area and there is no corresponding road in the first road area, when it is determined that the difference information is caused by road shielding (B1), the map information update unit 31 may determine not to update the map information. On the other hand, when it is not determined that the difference information is caused by road shielding (B0), it may be determined to update the map information. By doing so, the map information can be appropriately updated considering the presence or absence of shielding.

[0062] Also, when there are roads corresponding to both the first road area and the second road area (C1 and D1), the map information update unit 31 determines whether to update the map information (C1 or D1) based on the width of the road in the first road area and the width of the road in the second road area. By doing so, the map information can be appropriately updated considering the width of the road.

[0063] For E0 and F, no update of the reliability is required. Or in the case of E0, based on the verification result of the image information storage unit 10, information indicating that the reliability of the map information is high is given. Regarding A, it may be determined to update the map information, or information indicating that the map information is not updated because there is no road information to be updated on the map may be given. Or, regarding A, for example, at the intersection of a road with a difference and another road, information indicating low reliability is given. At this time, it is also conceivable to newly add nodes and links to the intersection in the network information and update it.

[0064] Next, consider the case of additionally registering in the map information whether the reliability of the number of lanes and the lane width is high or not. Here, when the reliability result of the difference amount in FIG. 6 is high, it is considered that the reliability of the information registered in the map is low, and the result is given. When the reliability of the difference amount in FIG. 6 is low, it is considered that the information registered in the map is reliable, and information indicating that the map information is not updated is given, or information indicating high reliability is given.

[0065] In the above, the mechanism for determining the reliability of the map information from the reliability of the difference has been described. Of course, the information of A to F may be directly registered as the information representing the reliability of the map information.

[0066] As described above, according to the information processing apparatus according to the present embodiment, the shielding area of the road can be appropriately handled. For example, by adopting a mechanism based on the presence or absence of the difference and the reliability of the output difference, it is possible to update the map information while considering the shielding information, which is a problem peculiar to satellite images and aerial photographs.

[0067] FIG. 3 shows a configuration in which a functional unit for estimating a shielding area is added to the first information conversion unit 15 and the second information conversion unit 17 shown in FIG. 2. The first information conversion unit 15 includes a first shielding area estimation unit 156, and the second information conversion unit 17 includes a second shielding area estimation unit 176.

[0068] The first information conversion unit 15 estimates first structure information based on an image by the first shielding area estimation unit 156. The first structure information is information for specifying, for example, a portion (shielding area) where road shielding exists in the image. The specific processing of the estimation can be appropriately designed based on known techniques or the like. When performing the estimation, the pre-trained model 150 may estimate a road area shielded from surrounding building groups and pattern information.

[0069] In the shielding area, there are cases where it is shielded by structures other than roads such as buildings and trees, and cases where it is shielded by another road such as an overpass or an expressway. Therefore, multiple attributes may be output for one pixel. For example, the attributes of structures other than roads and the attributes of the shielding road may be output, or the attributes of the road and the attributes of the shielding road may be output.

[0070] In addition, an output for estimating the cause of shielding such as shielding by a building, shielding by a tree, or shielding by another road may be provided.

[0071] The second information conversion unit 17 estimates second structure information based on map information by the second shielding area estimation unit ********. The second structure information is information for specifying, for example, a portion where road shielding exists in the map information. This estimation may be performed by the same processing as that of the map information verification unit 20. In the difference extraction unit 16, in this embodiment, the difference in the road area is obtained instead of the difference in the shielding area.

[0072] There are some differences between the shielding areas estimated by the first shielding area estimation unit 156 and the second shielding area estimation unit 176. One is that the shielding area that can be estimated by the shielding area estimation unit of the first information conversion unit 15 tends to be a shorter distance or a narrower area than that estimated by the second information conversion unit 17. For example, the first information conversion unit 15 can estimate shielding for a road where only a small part of the road is shielded, but cannot appropriately estimate shielding for an area where most of the road is shielded like a long tunnel.

[0073] In the configuration including the first shielding area estimation unit 156 and the second shielding area estimation unit 176 in this way, the configuration of the structure information verification unit 18 includes, for example, the mutual shielding information verification unit 21 in FIG. 8. In the mutual shielding information verification unit 21, the shielding situation is estimated by collating the first structure information and the second structure information.

[0074] In the mutual shielding information verification unit 21, if there is shielding in both the first structure information (image) and the second structure information (map information), 3 is output as the variable "$$" in FIG. 8; if there is shielding only in the map information, 2 is output; if there is shielding only in the image, 1 is output; and if there is no shielding in both, 0 is output. The meaning of the variable $$ is the same as that of the variable # in FIG. 5, but the variable $$ can take four values from 0 to 3. According to such a configuration, the shielding of each of the image and the map information can be considered.

[0075] In view of such a feature, the structure information verification unit 18 verifies the reliability of the extracted difference area. As a specific example, when a difference is extracted in the difference extraction unit 16, the shielding possibility is judged. If the whole of one road is shielded, the map information is not updated. Also, if only a part is shielded, the reliability of the difference is lowered. On the other hand, if it is judged that there is no shielding, the reliability of the difference is increased.

[0076] The above classifications A to F classify, in addition to whether the extracted difference is the presence or absence of a road, whether the reliability of the extracted difference is high and whether the amount of the difference is reliable on the premise that there is shielding in the image information storage unit 10.

[0077] Similar to FIG. 6, the reliability can be determined as shown in FIG. 9 for the configuration in FIG. 8. Regarding A, since there is no corresponding shielding information on the map side, A3 and A2 do not exist. Similarly, for F, F3 and F2 do not exist. The result classified in this way may be used as the output of the difference reliability output unit 30 shown in FIG. 10.

[0078] Based on the reliability shown in FIG. 9, the map information may be updated. The specific processing at this time can be designed in the same way as the example described in relation to FIG. 6.

[0079] After being updated with the configuration of FIG. 7, only the information in the image information storage unit 10 is updated, and an example in the case of collation will be described with reference to FIG. 12. Here, it is assumed that the reliability of the map information takes values from 0 to 3, and the larger the value, the higher the reliability of the map information. Note that this "reliability of the map information" is different information from the above-mentioned "reliability of the difference".

[0080] Here, it is considered that the data (image) in the image information storage unit 10 is newer than the data (map information) in the map information storage unit 11, but the following processing is equally applicable even if this is not the case.

[0081] When it is determined that there is no difference (S11: No), the map information update unit 31 sets the reliability of the map information to 3. When it is determined that there is a difference (S11: Yes), the following processing is performed.

[0082] When it is determined that the reliability of the presence or absence of the difference is low (S12: No), the map information update unit 31 sets the reliability of the map information to 2. When it is determined that the reliability of the presence or absence of the difference is high (S12: Yes), the following processing is performed.

[0083] The map information update unit 31 refers to the previous reliability (that is, the reliability of the map information determined when the map information was updated last time). When the previous reliability is not specified (S13: No), the reliability of the map information is set to 1. When the previous reliability is 0 or 1 (S14: Yes), the reliability of the map information is set to 0, and when the previous reliability is 2 or 3 (S14: No), the reliability of the map information is set to 1.

[0084] According to this example, the map information update unit 31 determines the reliability of the map information when updating the map information. Then, based on the reliability of the map information determined when the map information was last updated, it determines the new reliability of the map information. Therefore, based on the difference reliability, the reliability of the map information, and the history of the reliability of the map information, the reliability of the map information can be determined more appropriately. Since the reliability of the map information is used, for example, as a criterion for determining whether to update the map information, the map information can be updated more appropriately.

[0085] Next, an example in which the data (map information) in the map information storage unit 11 is newer than the data (image) in the image information storage unit 10 will be described. The update of the map information may be performed only on the nodes and links where differences have occurred from the previous map information. In that case, it is conceivable to discard the previous difference information related to the updated nodes and links and consider only the results of the newly performed collation.

[0086] Finally, when the newness of both data is the same (for example, when they are updated simultaneously), for the nodes and links where the map information is updated, the previous difference information is discarded and the new collation results are registered, and for the other parts, it is conceivable to apply the update rules shown in FIG. 12.

Explanation of Reference Numerals

[0087] 15…First information conversion unit, 16…Difference extraction unit, 17…Second information conversion unit, 18…Structure information verification unit, 30…Difference reliability output unit, 31…Map information update unit

Claims

1. A first information conversion unit that estimates a first road area from an image of a road; A second information conversion unit that estimates a second road area from map information stored in a map information storage unit; and a difference extraction unit that extracts difference information representing a difference between the first road area and the second road area; A difference reliability output unit that determines the reliability of the difference information; A map information update unit that determines whether to update the map information based on the difference information, comprising: The difference information includes difference presence / absence information indicating the presence or absence of a difference and difference amount information indicating the amount of the difference; The difference reliability output unit outputs a reliability for each of the difference presence / absence information and the difference amount information; The map information update unit determines whether to update the map information based on the reliability of the difference presence / absence information and the reliability of the difference amount information; The difference is a difference in road width or a difference in lane width; The difference reliability output unit determines the reliability based on a predefined correspondence relationship between a combination of whether there is the difference, whether there is a road in the first road area, whether there is a road in the second road area, and whether there is road occlusion, and the reliability; An information processing apparatus.

2. A first information conversion unit that estimates a first road area from an image of a road; A second information conversion unit that estimates a second road area from map information stored in a map information storage unit; and a difference extraction unit that extracts difference information representing a difference between the first road area and the second road area; A map information update unit that determines whether to update the map information based on the difference information, comprising: The map information update unit: In a case where there is a road in the second road area and there is no road corresponding to the first road area, Determines not to update the map information when it is determined that the difference information is caused by road occlusion; Determines to update the map information when it is not determined that the difference information is caused by road occlusion; An information processing apparatus.

3. A first information conversion unit that estimates a first road area from an image of a road; A second information conversion unit that estimates a second road area from map information stored in a map information storage unit; and a difference extraction unit that extracts difference information representing a difference between the first road area and the second road area; A map information update unit that determines whether to update the map information based on the difference information, comprising: When there is a road corresponding to both the first road area and the second road area, the map information updating unit further determines whether to update the map information based on the width of the road in the first road area and the width of the road in the second road area. When the width of the road in the first road area is less than or equal to the width of the road in the second road area, the map information updating unit determines not to update the map information. When the width of the road in the first road area is greater than the width of the road in the second road area, the map information updating unit determines to update the map information. Information processing apparatus.

4. An information processing apparatus according to any one of claims 1 to 3, wherein the map information updating unit determines the reliability of the map information when updating the map information. Information processing apparatus.

5. An information processing apparatus according to claim 4, wherein the map information updating unit determines the reliability of the map information based on the reliability of the map information determined when the map information was last updated. Information processing apparatus.

6. An information processing apparatus according to any one of claims 1 to 3, wherein the first information conversion unit estimates first structure information representing the structure of the road in the first road area based on the image, the second information conversion unit estimates second structure information representing the structure of the road in the second road area based on the map information, and the information processing apparatus includes a structure information verification unit that collates the first structure information and the second structure information. Information processing apparatus.

7. An information processing apparatus according to any one of claims 1 to 3, wherein when there is a road in the first road area and there is no road corresponding to the second road area, the map information updating unit determines to update the map information. Information processing apparatus.

8. A computer estimates a first road area from an image of a road, estimates a second road area from map information, extracts difference information representing the difference between the first road area and the second road area, determines the reliability of the difference information, determines whether to update the map information based on the difference information, and has wherein the difference information includes difference presence / absence information representing the presence or absence of a difference and difference amount information representing the amount of the difference, and in the step of determining the reliability, reliability is output for each of the difference presence / absence information and the difference amount information. In the step of determining whether to update the map information, based on the reliability of the difference presence / absence information and the reliability of the difference amount information, it is determined whether to update the map information. The difference is a difference in road width or a difference in lane width. In the step of determining the reliability, based on a predefined correspondence relationship between a combination of whether there is such a difference, whether there is a road in the first road area, whether there is a road in the second road area, and whether there is road occlusion, and the reliability, the reliability is determined. Information processing method.

9. A computer Estimating a first road area from an image obtained by imaging a road; Estimating a second road area from map information; Extracting difference information representing a difference between the first road area and the second road area; Determining whether to update the map information based on the difference information; and In the step of determining whether to update the map information, when there is a road in the second road area and there is no corresponding road in the first road area, if it is determined that the difference information is caused by road occlusion, it is determined not to update the map information; if it is not determined that the difference information is caused by road occlusion, it is determined to update the map information. Information processing method.

10. A computer Estimating a first road area from an image obtained by imaging a road; Estimating a second road area from map information; Extracting difference information representing a difference between the first road area and the second road area; Determining whether to update the map information based on the difference information; and In the step of determining whether to update the map information, when there are roads corresponding to both the first road area and the second road area, further based on the road width in the first road area and the road width in the second road area, it is determined whether to update the map information. In the step of determining whether to update the map information, when the road width in the first road area is less than or equal to the road width in the second road area, it is determined not to update the map information. In the step of determining whether to update the map information, when the width of the road in the first road area is larger than the width of the road in the second road area, it is determined to update the map information. Information processing method.

Images