JP7171666B2 - Road surface condition estimation device, road surface condition estimation program - Google Patents

Description

The present invention relates to a road surface condition estimation device and a road surface condition estimation program.

Traffic infrastructure centered on roads has deteriorated due to snow damage, flood damage, and deterioration of underground facilities such as sewage pipes, etc., resulting in holes, unevenness, steps, etc. . This has resulted in direct/indirect accidents involving automobiles, bicycles and pedestrians, increasing the burden on road administrators in terms of management responsibility and compensation.

Currently, road patrols and information provided by residents are the means to deal with this. Road patrols consist of walking patrols by staff members, visual inspections by dedicated vehicles, and patrols using dedicated equipment (see Patent Literatures 1 and 2).

Monitoring by road patrols is subject to restrictions on patrol intervals and the effects of other vehicle traffic due to issues such as cost.

JP 2020-013537 A JP 2019-109293 A

The prior art assumes the use of dedicated vehicles for detecting road surface conditions. This is advantageous in terms of accuracy in terms of visual inspection by a specialized person in charge, but depending on the time interval of confirmation and traffic volume, it is conceivable that confirmation may be difficult due to obstruction by other vehicles.

In addition, information provided by residents is reported only when a visible event occurs, and the accuracy and frequency of the content depends on the judgment of individual residents. The reliability of the detection accuracy was not sufficient.

The problem to be solved by the present invention is a road surface condition estimation device and a road surface condition estimation program that can detect road surface conditions with high reliability without patroling the road using a dedicated vehicle for detecting the road surface conditions. is to provide

According to the embodiment, the road surface condition estimating device has receiving means, recognizing means, and estimating means. The receiving means receives images of vehicles traveling on the road which are continuously captured. The recognition means recognizes the light of the vehicle and the headlamp from the image. The estimating means estimates the road surface condition of the road based on the positions of the plurality of vehicles recognized from the images and the change in the light corresponding to each vehicle, and the same vehicle recognized from the continuous images. An average value of a plurality of vehicles is calculated for the maximum fluctuation width of the brightness of the light corresponding to , and the road surface condition of the road is estimated based on the average value .

1 is a block diagram showing the configuration of a road surface condition estimation system according to this embodiment; FIG. FIG. 2 is a block diagram showing the configuration of a server (road surface condition estimation device) according to the present embodiment; FIG. 2 is a block diagram showing the functional configuration of a server according to the embodiment; FIG. BRIEF DESCRIPTION OF THE DRAWINGS The figure for demonstrating the outline|summary of the road surface condition estimation system in this embodiment. 4 is a flowchart for explaining recording processing of the server (road surface condition estimation device) according to the present embodiment; 4 is a flowchart for explaining the operation of server determination processing according to the present embodiment. FIG. 4 is a diagram for explaining the vibration of the vehicle caused by a damaged portion of the road surface; The figure which shows an example of the data of the vehicle vibration estimation result in this embodiment. The figure which shows an example of the data of the vehicle vibration statistical processing result in this embodiment. The figure which shows an example of the confirmation screen provided by the estimation result output part in this embodiment. The figure which shows an example of the detailed screen provided by the estimation result output part in this embodiment. FIG. 10 is a diagram showing an image recording state in a modified example of the present embodiment;

Hereinafter, embodiments will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a road surface condition estimation system according to this embodiment. The road surface condition estimation system is a system used to inspect the road surface conditions of roads (including tunnels, bridges, etc.). The road surface condition estimation system of this embodiment receives image data from a camera that captures a vehicle passing on the road and is installed around the road without using a dedicated vehicle for road surface monitoring, and processes the image data to detect the road surface. Estimate the situation. That is, in the road surface condition estimation system of the present embodiment, fixed cameras installed around the road, such as traffic lights, pedestrian bridges, signs, etc., or road surface conditions can be monitored without implementing road patrols using dedicated vehicles. We use images taken by cameras installed around the road. The road surface conditions include, for example, holes, unevenness, ruts, joints of bridges, steps at manholes, etc. (hereinafter referred to as damaged portions).

As shown in FIG. 1, the road surface condition estimation system in this embodiment includes a server 12, a plurality of cameras 10 (10-1, . . . , 10-m), electronic devices 16, traffic condition monitoring A system 18 is connected and configured. The server 12 in this embodiment operates as a road surface condition estimation device by a function realized by a road surface condition estimation program.

The server 12 (road surface condition estimating device) receives images of vehicles traveling on the road continuously photographed by the camera 10, and detects changes in the light of the vehicle's headlights and the characteristics of the light. The road surface condition is estimated based on the position of the vehicle when the change is detected, and the estimated result is output. That is, when there are no damaged parts such as holes, bumps, ruts, and steps on the road surface, the light from the vehicle headlights detected from the continuously captured images stabilizes as the vehicle approaches the camera 10. and change. On the other hand, when there is a damaged portion on the road surface, the position of the vehicle body fluctuates when the tires of the vehicle pass over the damaged portion. In general, a vehicle headlight can be considered to have an optical axis fixed to the vehicle body. Therefore, the change due to the light of the headlight (optical axis oscillation) may be regarded as the oscillation of the vehicle body. The server 12 estimates the road surface condition based on the position of the vehicle recognized from the image and changes in the light of the headlights.

As the camera 10, a fixed camera installed for the purpose of traffic congestion monitoring, accident monitoring, etc. on a road, which is installed at a traffic light, a pedestrian bridge, a sign, etc., can be used. In addition, the cameras 10 are installed around roads to be monitored for road surface conditions.

The traffic condition monitoring system 18 is, for example, a VICS (registered trademark) (Vehicle Information and Communication System) or a system that collects traffic condition information 18A regarding other road conditions and provides the traffic condition information. The traffic condition monitoring system 18 collects, for example, traffic condition information 18A such as information such as the number of passing vehicles, vehicle types, etc., and conditions such as traffic jams caused by accidents/construction that are different from usual. The server 12 receives the traffic condition information 18A from the traffic condition monitoring system 18 through the network 19, and uses it to determine reception/discard of the image received from the camera 10, and the like.

The electronic device 16 is, for example, a device such as a personal computer, a tablet computer, or a smart phone, and is used to access the server 12 and output the road surface condition estimation processing result. The electronic device 16 is used by an operator such as an infrastructure manager (road management company manager, local government manager, etc.). The server 12 outputs the processing result for road surface condition estimation on the display device provided in the server 12 and also outputs it on the electronic device 16 via the network 19 .

The network 19 includes various communication networks such as a wireless communication network, a wired communication network, and the Internet.

Note that the server 12 is realized by, for example, cloud computing, and may be realized by one server connected via a network 19 (Internet), or by a plurality of servers working together. good. In the following explanation, the server 12 is assumed to be realized by one server.

FIG. 2 is a block diagram showing the configuration of the server 12 (road surface condition estimation device) in this embodiment. The server 12 has a processor 12a, a memory 12b, a storage device 12c, a display device 12e, an input device 12f, and a communication device 12h.

The processor 12a is a circuit for executing a basic program (OS) and application programs stored in the memory 12b to realize various functions. For example, the processor 12a implements each functional unit described later by executing a road surface condition estimation program (see FIG. 3).

The memory 12b stores programs executed by the processor 12a, temporary data, and the like.

The storage device 12c stores various programs and various data. The data stored in the storage device 12c includes image data of images (or videos) received from the camera 10, traffic information received from the traffic monitoring system 18, map data, data of estimation processing results, and the like.

The display device 12e is an LCD (Liquid Crystal Display) or the like, and displays a screen according to the processing of the processor 12a. The input device 12f is a keyboard, pointing device, or the like.

The communication device 12h controls communication with external devices such as the camera 10, the traffic monitoring system 18, and other information processing devices through the network 19. FIG.

FIG. 3 is a block diagram showing a functional configuration 20 realized by executing a road surface condition estimation program by the processor 12a of the server 12 in this embodiment.

As shown in FIG. 3, the functional configuration 20 of the server 12 includes an image data reception processing unit 21, an image database (DB) 22, a traffic condition information reception unit 23, a vehicle vibration estimation unit 25, an information accumulation processing unit 27, a vehicle vibration An estimation result database (DB) 28 , a vehicle vibration statistical processing unit 29 , an estimation result output unit 30 and a map database (DB) 31 are included.

The image data reception processing unit 21 receives image data from the camera 10 through the network 19 . In this embodiment, the presence of a damaged portion on the road surface is estimated based on the light from the headlights of the vehicle captured in the image. Therefore, it is assumed that the image data reception processing unit 21 receives image data captured at night from sunset to dawn, for example. The image data reception processing unit 21 receives image data of continuous images cut out at regular intervals (for example, 0.05 seconds) from the video captured by the camera 10 .

Further, the image data reception processing unit 21 discards image data unsuitable for road surface condition estimation processing based on the traffic condition information 18A received from the traffic condition monitoring system 18 by the traffic condition information reception unit 23. can be For example, when receiving the traffic condition information 18A indicating that the road being photographed by the camera 10 is congested (or that an accident has occurred/construction is being carried out), the vehicle may travel differently than usual due to the occurrence of traffic congestion. Since there is a possibility that the vehicle is moving (low speed driving, etc.), the image data that is received when a traffic jam occurs is discarded. In addition, the image data may be discarded when the number of passing vehicles is significantly different from normal, or when the types of passing vehicles are different (such as when there are many large vehicles such as trucks on a road where many passenger cars normally pass). In addition, the image data may be discarded depending on the weather conditions (heavy rain or fog).

The image DB 22 stores image data received by the image data reception processing unit 21 .

The traffic condition information receiving unit 23 receives the traffic condition information 18A from the traffic condition monitoring system 18 and stores it in the storage device 12c.

The vehicle motion estimation unit 25 recognizes the light of the headlights of the vehicle from the images of the vehicle traveling on the road, which are continuously captured by the camera 10 and stored in the image DB 22, and detects the headlights of the vehicle. and the position of the vehicle when the characteristic fluctuation of the light (for example, the maximum fluctuation width of the light intensity of the headlamp) is detected, the road surface condition is estimated based on the estimation result of the vehicle vibration.

The vehicle vibration estimation unit 25 estimates road surface conditions based on the positions of a plurality of vehicles recognized from the images stored in the image DB 22 and changes in the light of the headlamps corresponding to each vehicle, and obtains the estimation result. A recording process is executed to generate and record the vehicle vibration estimation result indicating the .

The information accumulation processing unit 27 causes the vehicle vibration estimation result DB 28 to store the vehicle vibration estimation result for each vehicle obtained by the vehicle vibration estimation unit 25 .

A vehicle vibration statistical processing unit 29 calculates statistical values for determining the road surface condition based on the vehicle vibration estimation results corresponding to a plurality of vehicles stored in the vehicle vibration estimation result DB 28, and determines the road surface damage. determination processing for determining the occurrence of That is, the vehicle vibration statistical processing unit 29 obtains statistical values representing changes in the light of the headlights for a plurality of vehicles, and estimates changes in the damaged portions of the road surface from changes in the statistical values over time. .

The estimation result output unit 30 outputs the determination result (information for notifying the state of the road surface) by the vehicle vibration estimation unit 25 or the vehicle vibration statistical processing unit 29 . The estimation result output unit 30 outputs the determination result to the display device 12e of its own device, the electronic device 16 connected via the network 19, or the like. The display screen as the determination result includes, for example, a map showing the location (or road, photographing range of the camera, etc.) determined to be the damaged portion based on the map data recorded in the map DB 31, and the vehicle vibration statistical processing unit 29. A graph showing changes in the statistical values obtained by , shooting information (installation position, information on the camera, etc.) related to the camera 10 that captures a portion determined to be a damaged portion, etc. are displayed. In addition, the estimation result output unit 30 also provides information about roads that have been determined to be damaged by the determination processing by the vehicle vibration statistical processing unit 29, in order to facilitate the work of confirming the damaged area by the operator (infrastructure manager, etc.) for the determination result. A display screen for displaying a list of information may be output.

The map DB 31 is a database in which map data is stored, and includes position data indicating routes for each road.

Next, the operation of the road surface condition estimation system according to this embodiment will be described.

FIG. 4 is a diagram for explaining the outline of the road surface condition estimation system in this embodiment. As shown in image recording situation 2 in FIG. 4, while the vehicle is traveling on the road, a fixed camera 10 installed, for example, at a traffic light continuously photographs the vehicle passing on the road. The camera 10 may photograph not only the passage of one vehicle within the photographing range, but also the passage of a plurality of vehicles. Images captured by camera 10 are collected at server 12 .

The server 12 can collect images from the cameras 10 installed around a number of roads, and estimate the occurrence of damaged portions in the road surface conditions of the roads to be photographed by the respective cameras 10 . Based on the traffic condition information 18A provided from the traffic condition monitoring system 18, the server 12 discards image data unsuitable for road surface condition estimation processing, so that determination accuracy can be maintained. By estimating the occurrence of damaged parts based on images collected from a large number of cameras 10, the server 12 can monitor the road surface conditions in the same way as when many roads are patrolled frequently. .

As shown in confirmation status 4 shown in FIG. 4 , an operator such as an infrastructure administrator (road management company person in charge, local government person in charge, etc.) can confirm the processing result by the server 12 through the electronic device 16 . In the electronic device 16, the position determined as the damaged part by the server 12 is clearly indicated on the map, and information about the damaged part (installation location of the camera 10, road position, etc.) can be confirmed, so road surface conditions can be easily confirmed. can do.

FIG. 5 is a flowchart for explaining the recording process of the server 12 (road surface condition estimation device) in this embodiment, and FIG. 6 is a flowchart for explaining the determination process operation of the server 12 in this embodiment.

The image data reception processing unit 21 (processor 12a) of the server 12 receives image data from the camera 10 via the network 19 by image data processing (step A1). It is assumed that the image data is added with shooting information such as shooting time (shooting time).

For example, the image data reception processing unit 21 receives image data captured by one camera 10 at night from sunset to dawn. For example, the time period for capturing images to be provided to the server 12 may be set for each camera 10, not limited to the period from sunset to dawn. In this case, the target time period for each camera 10 can be set for the server 12 by an administrator's operation or the like.

Depending on the road, there are cases where the proportion of vehicle types that travel on the road is stable during a specific time period. For example, there are times when the proportion of large vehicles (trucks) is large, or times when the proportion of large vehicles and passenger cars is almost constant. By performing determination processing based on images captured during such time periods, stable statistical values can be easily calculated, and the accuracy of estimating damaged portions on the road surface can be improved.

The image data reception processing unit 21 determines whether to discard/record the image data received from the camera 10 based on the traffic condition information 18A received from the traffic condition monitoring system 18 by the traffic condition information reception unit 23 (step A2). ). Among the image data received from the camera 10, the image data reception processing unit 21 discards the image data in the time zone in which an image inappropriate for road surface condition estimation processing may have been captured, and discards the other image data as an image. Record in DB 22 (step A3).

Next, in the vehicle motion estimation process, the vehicle motion estimation unit 25 executes processing for estimating vehicle motion for each image captured by one camera 10 (each road monitored by the camera 10).

The vehicle motion estimator 25 recognizes the vehicle captured in the image and the light from the vehicle's headlights using, for example, a known object recognition processing technique, and tracks the motion within a plurality of consecutive images. The vehicle motion estimation unit 25 sets a unique identifier (vehicle ID) for the vehicle recognized from the image. In addition, the vehicle motion estimation unit 25 sets an identifier (image ID) for each of the plurality of images used for tracking the vehicle and the light from the headlights. When a plurality of vehicles are photographed in the image, the light of each vehicle and its headlight is individually recognized to track the movement (step A4).

Further, the vehicle motion estimation unit 25 estimates the light of the headlights of the vehicle, here, the brightness of the headlights in the image (step A5). If multiple vehicles are captured in the image, each vehicle is individually tracked and the brightness of the headlights of each vehicle is estimated.

For each vehicle extracted from the image, the vehicle vibration estimation unit 25 records changes in the brightness of the headlights as fluctuation width time-series data, and calculates the maximum fluctuation width based on the fluctuation width time-series data ( Step A6).

For example, as shown in FIG. 7A, when there is no damaged portion on the road surface, the vehicle CA1 traveling on the road is less likely to sway. In this case, since the headlight of the vehicle generally has an optical axis fixed to the vehicle body, the fluctuation of the optical axis is also reduced. Therefore, there is little change in the brightness of the headlights recognized from the image captured by the camera 10, and the change is stable according to the passage of vehicles.

On the other hand, as shown in FIG. 7(B), when there is a damaged portion (for example, a convex portion) on the road surface, the vibration of the vehicle body increases when the vehicle CA2 passes through the damaged portion (convex portion), causing the headlights to sway. The optical axis of the lamp also changes. In this case, the change in the brightness of the headlight location recognized from the image captured by the camera 10 also increases when the vehicle passes through the damaged portion (for example, a convex portion).

Even if the damaged portion on the road surface is small and the vehicle body of the vehicle CA2 passing through the damaged portion is shaken little, the image captured by the camera 10 installed at a position away from the vehicle CA2 shows that the headlights light fluctuates greatly. Therefore, it is possible to detect the occurrence of damaged parts based on changes in the light of the vehicle's headlights.

Further, the vehicle vibration estimation unit 25 calculates the position on the road of the vehicle recognized from the image based on the image for which the maximum fluctuation range has been calculated. For example, for the camera 10 that captured the image to be processed, the camera angle of view information (camera orientation angle, camera installation height, shooting range, relative position to the road, etc.), camera attributes (camera type, shooting conditions (aperture etc.) is known. The server 12 records camera attribute information for each camera 10 in advance. The vehicle motion estimation unit 25 calculates the position of the vehicle on the road based on the position of the vehicle recognized from the image and the camera attribute information.

It should be noted that the position of the vehicle on the road obtained by the vehicle motion estimation process does not need to be a precise position. That is, when it is determined that there is a damaged portion on the road surface, it is sufficient if the periphery of the damaged portion can be identified because the administrator actually checks it by visual inspection or the like. For example, the range of the road to be photographed by the camera 10 is divided into a plurality of road sections, and any road section that includes the vehicle is specified as the position of the vehicle on the road.

The vehicle motion estimator 25 creates a vehicle motion estimation result including the information obtained by the processes described above.

The information accumulation processing unit 27 records the vehicle vibration estimation result created by the vehicle vibration estimation unit 25 in the vehicle vibration estimation result DB 28 (step A8).

FIG. 8 shows an example of vehicle vibration estimation result data in this embodiment.

As the data of the vehicle vibration estimation result, for each vehicle recognized from the image, for example, date/time, vehicle ID, image ID string, image position, target vehicle, maximum amplitude fluctuation width, and fluctuation width time series are associated. Recorded. Date/Time indicates the date and time when the image in which the vehicle was recognized was taken. A vehicle ID is a unique identifier set for a vehicle recognized from an image. The image ID column is an image ID list of identifiers (image IDs) set for each of a plurality of images used for tracking the light of the vehicle and headlights. The image position is information indicating the position (for example, road segment) of the target image. The target vehicle is a list of positions of vehicles specified in each image for each image ID for vehicles recognized from the image. The maximum amplitude fluctuation width indicates the fluctuation width of the maximum amplitude of the headlamp light quantity corresponding to the vehicle recognized from the image. The variation width time series is variation width time series data in which changes in brightness of the headlights of the target vehicle are recorded in time series.

Next, determination processing of the server 12 will be described.

In the vehicle motion statistical processing, the vehicle motion statistical processing unit 29 of the server 12 selects the road to be processed (where the camera 10 is installed) based on the vehicle motion estimation results of a plurality of vehicles recorded in the vehicle motion estimation result DB 28. Statistical values for judging the road surface condition are calculated for each position), and the occurrence of a damaged portion of the road surface is judged. It is assumed that the vehicle vibration statistical processing is executed at a predetermined statistical processing execution cycle, for example, once a month.

First, in the vehicle motion statistical processing, the vehicle motion statistical processing unit 29 uses a preset average time width, for example, the maximum An amplitude fluctuation range average value and a maximum amplitude fluctuation range variance value are calculated (step B1). That is, the vehicle vibration statistical processing unit 29 calculates the daily average maximum amplitude fluctuation range for all vehicles recognized from the images captured by the camera 10 within one day, as shown in the following formula (1). As in (2), the daily maximum amplitude fluctuation width dispersion value is calculated.

Average maximum amplitude fluctuation range (date)
= (Σ (maximum fluctuation width (vehicle ID)) / (total number of vehicles) (1)
Maximum amplitude variation range variance (date)
= (Σ (maximum fluctuation width (vehicle ID) - maximum amplitude fluctuation width average value (date)) ² ) / (total number of vehicles) (2)
Furthermore, the vehicle vibration statistical processing unit 29 calculates the maximum amplitude fluctuation width average value change rate and the maximum amplitude fluctuation width variance value change rate within a preset statistical processing execution period, for example, within one month, using the following formula (3 ) to create vehicle vibration statistical processing results (step B2).

Fluctuation width change rate (date) = Abs (1 - ((maximum amplitude fluctuation width average value (date)) / (Σ (maximum amplitude fluctuation width average value (date)) / (statistical processing execution cycle width))) ... ( 3)
The vehicle vibration statistical processing unit 29 determines whether the maximum amplitude fluctuation width average value change rate and the maximum amplitude fluctuation width variance value change rate within the statistical processing execution period (within one month) are included in any of the preset reference ranges. The road surface condition is determined depending on whether the road surface is clear or not (step B3).

For example, the vehicle vibration statistical processing unit 29 determines that the road surface condition is normal when it is determined that the rate of change of the maximum amplitude fluctuation range average value for each date is within 10% of the average within one month. If it is in the range of 10% or more and less than 20% of the average, it is judged to be in a caution state, and if it is in the range of 20% or more, it is judged to be in an abnormal state.

Similarly, both fluctuation statistical processing section 29 determines the rate of change of maximum amplitude fluctuation range dispersion value for each date. The reference range for the maximum amplitude fluctuation width variance value change rate may be the same as or different from the maximum amplitude fluctuation width mean value change rate.

Also, in the above description, determination is made in three stages of normal state, caution state, and abnormal state, but determination may be made in several other stages. In that case, the reference range shall be set according to each stage.

In addition, a different value may be set for the reference range for each road to be processed (for each camera 10 that captured the image to be processed). For example, for roads that can tolerate damage, the reference value (20% in the example described above) for determining abnormal conditions is increased, and conversely, for roads that require attention even for slight damage, caution is applied. It is also possible to lower the reference value for judging the abnormal state.

The vehicle vibration statistical processing unit 29 records each data of the vehicle vibration statistical processing result in the storage device 12c.

FIG. 9 shows an example of data resulting from statistical processing of vehicle vibration in this embodiment.

As the data of the vehicle vibration statistical processing result, the judgment date/time, the position information, the estimation result ID string, the judgment result, the parameters for statistical processing, the variation width time series, and the dispersion time series are recorded in association with each other. The date/time indicates the date and time of the oldest data of the vehicle vibration estimation results used for processing. The position information is information indicating the information (for example, road segment) of the image position to be processed. The estimation result ID column indicates an identifier list of estimation result IDs attached to individual data of vehicle vibration estimation results used for processing. The determination result indicates the determination result (normal state/caution state/abnormal state) determined based on the statistical value. The statistical processing parameter indicates the time width for statistical processing and the statistical processing execution cycle. The variation width time series indicates time series data within the statistical processing cycle of values obtained by averaging the maximum amplitude variation widths of the motion estimation results for each time width. The variance time series indicates time series data within the statistical processing cycle of the variance value for each time width of the maximum amplitude fluctuation width of the motion estimation result.

The vibration of the vehicle body is affected by the vibration absorbing mechanism provided in the vehicle body of each vehicle, and changes depending on the size of the unevenness on the road surface caused by the damaged part and the size of the vehicle body. However, by obtaining statistical values for vehicle vibrations detected by vehicles traveling on roads based on vehicle vibration estimation results detected from a large number of vehicles, it is possible that these statistical values will not be affected by damaged parts. converge. In particular, in the case of general roads, if the number of passing vehicles and the type of vehicles do not change significantly within a period of one day, for example, it can be expected that the effect of the damaged portion will be remarkably expressed. Therefore, if the time span within which the number of passing vehicles and vehicle types are generally constant is used for calculating the average value of the maximum amplitude fluctuation range, it can be expected that the vehicle body vibration will reflect changes in the road surface such as bumps and dips.

In addition, when the road surface is partially uneven and the vehicles affected by the road surface are limited, such as vehicles whose tires run on the road surface, it can be expected that the dispersion value of the maximum amplitude fluctuation width will change.

When the server 12 receives a processing result display request from an operator such as an infrastructure administrator through the electronic device 16 , the server 12 provides the electronic device 16 with a screen showing the processing result. In the display process, the estimation result output unit 30 creates a confirmation screen D<b>1 that displays the determination result of the vehicle vibration statistical processing result and the position information using the map data of the map DB 31 , and causes the electronic device 16 to display the confirmation screen D<b>1 .

FIG. 10 is a diagram showing an example of the confirmation screen D1 provided by the estimation result output unit 30 in this embodiment.

In the confirmation screen D1 shown in FIG. 10, a position (range) D11 determined as a caution state or an abnormal state is displayed on a map according to the determination result (caution state, abnormal state) of the vehicle vibration statistical processing result and the position information. Here, for example, based on the position information, the position of the road section may be displayed with a mark or the like, or the range of the road portion of the range to be photographed by the camera 10 including the corresponding road section may be displayed. . In addition, by changing the display mode according to the determination result, such as yellow for positions determined to be in a caution state and red for positions determined to be in an abnormal state, the confirmation operation by the operator can be facilitated. can.

In the confirmation screen D1, when the location is clicked with the cursor D12 by the operation of the operator, the estimation result output unit 30 performs detailed calculation based on the data of the vehicle vibration statistical processing result corresponding to the position specified by the click operation. Display the screen D2. Here, as the detailed screen, for example, based on the variation width time-series data of the vehicle vibration statistical processing result, a graph showing the change in the value obtained by averaging the maximum amplitude variation width for each time width is displayed.

FIG. 11 is a diagram showing an example of the detailed screen D2 provided by the estimation result output unit 30 in this embodiment.

The detailed screen D2 shown in FIG. 11 displays a graph showing time-series data of fluctuation width, an average value, and reference values (10%, 20%) for judging the road surface condition. As a result, the operator can estimate how the state of the damaged part at the position to be checked changes over time.

In FIG. 11, the graph is displayed based on the time-series data of fluctuation width, but the graph may be displayed based on the data of dispersion time-series at the same time. Also, an image captured by the camera 10 at the corresponding position based on the position information may be displayed. Furthermore, a detailed screen can be created and displayed on the electronic device 16 based on the data recorded as other vehicle motion statistical processing results or vehicle motion estimation results, in addition to the fluctuation width time series and variance time series data. You can do it. By confirming the estimation result by the server 12 on the confirmation screen D1 and the detail screen D2, the operator can use it for work such as planning the next patrol by a dedicated vehicle or patrol route on foot, thereby improving work efficiency. can be achieved.

In this way, the server 12 (road surface condition estimation device) in this embodiment detects the shaking of the vehicle body caused by the influence of the damaged part of the road surface (unevenness, etc.) as a change in the amount of light of the headlights, and at a certain point, Statistical values are obtained for the magnitude of light intensity changes for each vehicle, for example, in the observation time span that can be regarded as the average number of passing vehicles and vehicle types, and from the changes in the statistical values over time, changes in the road surface over time are estimated. and the occurrence of damage can be determined. Therefore, the road surface condition can be detected with high reliability without patrol using a dedicated vehicle for detecting the road surface condition.

In the above description, the existence of the damaged portion is estimated from the image of the vehicle including the headlight portion photographed by the camera 10, but it is possible to carry out a modified example in which the headlight portion is not directly photographed. can. For example, a modification will be described in which a projection plate is installed on the side of a road or the like, and the fluctuation range of light projected onto this projection plate from a vehicle is monitored.

FIG. 12 is a diagram showing an image recording situation 6 in a modified example of this embodiment. As shown in image recording situation 6 in FIG. 12, a projection plate 40 installed on the side of the road in front of the vehicle in the traveling direction is photographed by a camera 10B installed on a traffic light or the like. Note that the camera 10B may capture the projection plate 40 at its angle of view. A traffic light or the like is provided with a camera 10A for photographing a vehicle traveling on the road.

The image captured by the camera 10A is processed to calculate the position of the vehicle on the road, as described above. On the other hand, for the image captured by the camera 10B, the vehicle vibration estimation processing is executed in the same manner as described above.

In the vehicle vibration estimation process, instead of detecting changes in the amount of light in the headlights of the vehicle photographed in the images, the fluctuation range in the vertical direction of the projected light in successive images for each vehicle is detected. Changes in the variation width of the projected light on the continuous images are recorded as time-series data during the time Ts acquired for each vehicle, and the maximum variation width at each time Ts is obtained as the maximum variation width (i), and the vehicle It records in vehicle vibration estimation result DB28 as a vibration estimation result.

At this time, the time Ts has the following values, for example, assuming that the effective range of the headlight for passing (downward) is 40 m and the average speed during night driving is 30 km/h.

Ts=40/(30000/3600)=4.8 [s] (4)
Also, in the measurement section corresponding to the time Ts, the vehicle approaching the projection plate 40 may be photographed by the camera 10A shown in FIG. 12 and controlled according to its position.

Subsequently, in the vehicle vibration statistical processing, from the vehicle vibration estimation results for each vehicle recorded in the vehicle vibration estimation result DB 28, the average time width, for example, all vehicles observed within one day (the number of acquired times Ts), The daily average maximum amplitude fluctuation range is calculated according to the following formula (5), and the daily maximum amplitude fluctuation range dispersion value is calculated according to the following formula (6).

Average maximum amplitude fluctuation range (date)
= (Σ maximum fluctuation range (i))/(number of Ts intervals acquired within one day) (5)
Maximum amplitude variation range variance (date)
= (Σ (maximum fluctuation width (i) - maximum amplitude fluctuation width average value (i)) ² )/(number of Ts intervals acquired within one day) (6)
In the vehicle vibration statistical processing, in the same manner as described above, the statistical processing execution cycle of the maximum amplitude fluctuation range average value (date) and the maximum amplitude fluctuation range variance value (date), for example, the rate of change within one month is calculated, and the rate of change is calculated. Based on this, the road surface condition is determined, and the estimated results are presented to the operator through display processing.

In the above description, it is assumed that the fluctuation range in the vertical direction of the projected light is detected. may be detected respectively.

In this way, by estimating the similarity of the vehicle based on the variation of the light projected onto the projection plate 40 from the headlights of the vehicle, even a slight vibration of the vehicle caused by a damaged portion of the road surface can be projected. Since the light projected onto the plate 40 fluctuates greatly, the vibration of the vehicle caused by the damaged portion of the road surface can be detected more clearly. Therefore, it is possible to reliably determine the damaged portion of the road surface.

It should be noted that the methods described in each of the above embodiments can be applied to programs that can be executed by a computer such as a magnetic disk (flexible disk, hard disk, etc.), an optical disk (CD-ROM, DVD, etc.), a magneto-optical disk (MO), It can also be stored in a storage medium such as a semiconductor memory and distributed.

Moreover, as long as the storage medium can store the program and is readable by the computer, the storage format may be any form.

In order to realize the above embodiments, the OS (operating system) running on the computer based on the instructions of the program installed in the computer from the storage medium, MW (middleware) such as database management software, network software, etc. You may perform a part of each process of .

Furthermore, the storage medium in each embodiment is not limited to a medium independent of a computer, and includes a storage medium in which a program transmitted via LAN, Internet, etc. is downloaded and stored or temporarily stored.

Moreover, the storage medium is not limited to one, and the storage medium in the present invention includes the case where the processing in each of the above embodiments is executed from a plurality of media, and the medium configuration may be any configuration.

The computer in each embodiment executes each process in each of the above embodiments based on a program stored in a storage medium. Any configuration, such as a connected system, may be used.

Further, the computer in each embodiment includes an arithmetic processing unit, a microcomputer, etc. included in information processing equipment, and collectively refers to equipment and devices capable of realizing the functions of the present invention by means of a program.

It should be noted that although several embodiments of the invention have been described, these embodiments are provided by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the scope of the invention described in the claims and equivalents thereof.

10 (10-1, ..., 10-m) ... camera, 12 ... server, 12a ... processor, 12b ... memory, 12c ... storage device, 12e ... display device, 12f ... input device, 12h ... communication device, 18 ... traffic Situation monitoring system.

Claims (8)

Receiving means for receiving continuously captured images of vehicles traveling on the road;
recognition means for recognizing the light of the vehicle and headlights from the image;
The road surface condition of the road is estimated based on the positions of a plurality of vehicles recognized from the images and the changes in the light corresponding to each of the vehicles. and an estimating means for calculating an average value of a plurality of vehicles with respect to the maximum fluctuation width of light brightness, and estimating the road surface condition of the road based on the average value . Receiving means for receiving continuously captured images of vehicles traveling on the road;
recognition means for recognizing the light of the vehicle and headlights from the image;
The road surface condition of the road is estimated based on the positions of a plurality of vehicles recognized from the images and the changes in the light corresponding to each of the vehicles. and an estimating means for calculating an average value of a plurality of vehicles with respect to the maximum variation width of light in the vertical direction, and estimating the road surface condition of the road based on the average value . a receiving means for receiving a first image continuously photographing a vehicle traveling on a road and a second image photographing a projection plate onto which the light of the headlight of the vehicle is projected;
recognition means for recognizing a vehicle from the first image and recognizing light projected onto the projection plate from the second image;
A road surface condition estimating device, comprising an estimating means for estimating a road surface condition of the road based on the position of the vehicle recognized from the first image and the change in the light. further comprising detecting means for detecting the position of the vehicle traveling on the road;
4. The receiving means according to any one of claims 1 to 3, wherein the receiving means receives an image taken when the vehicle is positioned within a shooting target range based on the position detected by the detecting means. road surface condition estimation device. the computer,
Receiving means for receiving continuously captured images of vehicles traveling on the road;
recognition means for recognizing the light of the vehicle and headlights from the image;
The road surface condition of the road is estimated based on the positions of a plurality of vehicles recognized from the images and the changes in the light corresponding to each of the vehicles. A road surface condition estimation program for calculating an average value of a plurality of vehicles with respect to the maximum fluctuation width of light brightness, and functioning as estimation means for estimating the road surface condition of the road based on the average value . the computer,
Receiving means for receiving continuously captured images of vehicles traveling on the road;
recognition means for recognizing the light of the vehicle and headlights from the image;
The road surface condition of the road is estimated based on the positions of a plurality of vehicles recognized from the images and the changes in the light corresponding to each of the vehicles. A road surface condition estimation program for calculating an average value of a plurality of vehicles with respect to the maximum variation width of light in the vertical direction, and functioning as estimation means for estimating the road surface condition of the road based on the average value. the computer,
a receiving means for receiving a first image continuously photographing a vehicle traveling on a road and a second image photographing a projection plate onto which the light of the headlight of the vehicle is projected;
recognition means for recognizing a vehicle from the first image and recognizing light projected onto the projection plate from the second image;
A road surface condition estimation program for functioning as estimation means for estimating the road surface condition of the road based on the position of the vehicle recognized from the first image and the change in the light. the computer,
Further functioning as a detection means for detecting the position of a vehicle traveling on the road,
Claims 5 to 5, wherein the receiving means functions to receive an image taken when the vehicle is positioned within a shooting target range based on the position detected by the detecting means. 8. The road surface condition estimation program according to any one of 7.

Images