JP7333511B2 - Road maintenance management system, pavement type determination device, repair priority determination device, road maintenance management method, pavement type determination method, repair priority determination method - Google Patents

Description

The embodiments of the present invention relate to a road maintenance management system, a pavement type determination device, a pavement deterioration determination device, a repair priority determination device, a road maintenance management method, a pavement type determination method, a pavement deterioration determination method, and a repair priority determination method.

Conventionally, road pavement inspections have been carried out to investigate road surface properties using dedicated measurement vehicles. On the other hand, when inspecting the vast number of roads managed by municipalities, conventional surveys are costly. Inspection technology using a simple method of image processing to determine road surface deterioration conditions such as cracks and peeling is being studied.

However, the surface of the road is not uniform, and the appearance of the road surface differs depending on the type of pavement, and the progress of deterioration also differs. In spite of this, deterioration is judged based on the same standard in the past, so there are cases where the inspection results differ from the actual situation even in the current road surface condition survey. In addition, since the type of pavement is not accurately managed, there are many cases where the actual situation is unknown or inconsistent. Accurate management costs a huge amount of money, so technology that can easily classify pavement types from road images and location information, determine the degree of deterioration for each type, and determine the priority of repairs. is required.

JP 2018-087484 A

The problems of the embodiments of the present invention are that it is possible to easily classify the type of pavement from road images and position information, determine the degree of deterioration for each type of pavement, and determine the priority of repair. To provide a road maintenance management system, a pavement type determination device, a pavement deterioration determination device, a repair priority determination device, a road maintenance management method, a pavement type determination method, a pavement deterioration determination method, and a repair priority determination method be.

According to an embodiment, a road maintenance management system includes a pavement type determination device, a pavement deterioration determination device , and a repair priority determination device . The pavement type determination device includes an input unit for inputting a road image obtained by photographing a road and photographing position information of the road image, an extraction unit for extracting an image of a road surface portion from the road image input to the input unit, Paving that determines the pavement type of the road surface portion of the road image using a plurality of learning models generated by learning the pavement type image of the road surface portion given for each pavement type with teacher data classified by attribute unit. and a type determination unit. The pavement deterioration determination device includes an image analysis unit that detects cracks or peeling of the pavement in the road surface portion of the road image whose pavement type has been determined by the pavement type determination device and analyzes the state of pavement deterioration; a pavement deterioration determination unit that determines the degree of pavement deterioration for each pavement type based on the detection result of the pavement crack or peeling location obtained by the analysis unit and the analysis result of the deterioration state of the pavement. The repair priority determination device receives input information of section registration data including the road image, the shooting position information, the pavement type determination result, the pavement deterioration determination result, and management information registered for each section of the road. Create a deterioration prediction formula that predicts the degree of future deterioration using regression analysis based on the data registration unit registered in correspondence, the pavement type determination result, the pavement deterioration determination result, and the section registration data. Based on the deterioration prediction formula creation unit, the judgment result of the pavement type, the judgment result of the pavement deterioration, the section registration data, and the prediction result by the deterioration prediction formula, repair priority is determined for each section of the road. and a priority determination unit for determining.

1 is a block diagram showing the overall configuration of a road maintenance system according to an embodiment; FIG. 2 is a diagram showing an example of a research vehicle used in the system shown in FIG. 1 and photographed road image data; FIG. 3 is a diagram showing types of asphalt mixtures as road pavement types to be managed in the system shown in FIG. 1; FIG. FIG. 4 is a block diagram showing the configuration of a pavement type determination device used in the system of the embodiment; 5 is a diagram for explaining a method of creating teacher data for AI applied to the pavement type determination apparatus shown in FIG. 4; FIG. 6 is a flowchart showing the flow of processing of the pavement type determination device shown in FIG. 4; FIG. 7 is a diagram showing a display example of a determination result by the pavement type determination process shown in FIG. 4; FIG. 8 is a block diagram showing the configuration of a pavement deterioration determining device used in the system shown in FIG. 1; FIG. 9 is a flowchart showing the flow of processing of the pavement deterioration determination device shown in FIG. 8; FIG. 10 is a diagram showing a display example of a determination result by the pavement deterioration determination process shown in FIG. 9; FIG. 11 is a block diagram showing a configuration of a repair priority determination device used in the system shown in FIG. 1; FIG. 12 is a flowchart showing the flow of processing of the repair priority determination device shown in FIG. 11; FIG. FIG. 13 is a diagram showing an example of a map display of determination results of the repair priority determination process shown in FIG. 12 ; 14 is a diagram showing a display example of a list (form) of determination results of the repair priority determination process shown in FIG. 12; FIG.

Various embodiments will be described below with reference to the drawings.
(First embodiment)
FIG. 1 is a block diagram showing a schematic configuration of a road management system according to the first embodiment. The road maintenance management system 1 shown in FIG. A pavement deterioration determination device 12 that automatically determines the degree of deterioration for each pavement type based on the pavement type determination result, and priority of repair based on road image data, camera position information, pavement type determination result, and pavement deterioration determination result. and a repair priority determination device 13 that determines the degree of repair.

FIG. 2 is a diagram showing an example of survey vehicles used in the system 1 shown in FIG. 1 and photographed road image data. A road image may be a moving image or a still image. As shown in FIG. 2(a), the system 1 uses a research vehicle 10 equipped with a camera 101 for photographing a road and a positioning device 102 for collecting position data of photographing. However, the vehicle itself need not be exclusive.

The camera 101 is mounted on the roof of the vehicle 100, for example, with the field of view directed so as to photograph the road behind the vehicle in the direction of travel. As a result, a road image as shown in FIG. 2(b) is obtained. The road image is an image taken obliquely from above, but an image taken from directly above may be used. The camera 101 adds information such as the orientation of the photographing direction, the viewing angle, the height from the road surface, the model, etc., to the photographed image data and outputs the information. The camera 101 may be a commercially available video camera or a smartphone camera.

Based on map data and latitude and longitude information obtained from a user receiver (GPS receiver) of a GPS (Global Positioning System) installed in the vehicle, the positioning device 102 detects the section of the route in synchronization with the photographing of the camera 101. Attribute information and position information of the camera at the time of shooting are collected and output as camera position information. The camera position information may include information such as altitude, speed, date and time, in addition to latitude and longitude information. The camera position information may be corrected using the speed information and acceleration information of the vehicle equipped with the camera in mountainous areas or tunnels where radio waves from GPS satellites are difficult to reach.

Here, the pavement of the above road can ensure high road surface performance, but due to the characteristics of the material used, there is a large variation in the progress rate of deterioration due to the characteristics of the material used. It is roughly classified into "concrete pavement" that can be expected to continue to be used over a period of time. Asphalt pavement can be broadly classified into two types depending on whether or not it has a structure that allows rainwater to smoothly permeate under the road surface. Of these, the one that uses an impermeable material such as a dense-grained asphalt mixture on the surface layer and drains water on the road surface is called "dense-grained asphalt pavement" and uses a porous asphalt mixture with water permeability on the surface layer or surface layer / base layer. This pavement is called "porous asphalt pavement". Concrete pavement, anti-flexible pavement, etc. are classified as "Other pavement".

FIG. 3 exemplifies types of asphalt mixtures for reference (for details, see the website of the Japan Asphalt Association (http://www.askyo.jp/knowledge/06-3.html)).
The surface of the road is not uniform, and the manner and progress of deterioration differ depending on the type of pavement. In spite of this, since the same criteria are used to determine deterioration, there are cases where the inspection results differ from the actual situation even in the current road surface condition surveys. Therefore, it is necessary to grasp the pavement type. Originally, the type of pavement should be managed in a ledger along with the construction history by each road administrator, but the reality is that the ledger is not satisfactorily maintained for some prefectural roads and most municipal roads. be. Therefore, we propose a method to automatically determine the pavement type based on images taken with a commercially available video camera.

FIG. 4 is a block diagram showing a specific configuration of the pavement type determination device 11 used in the road maintenance management system 1. As shown in FIG.
In the system 1 shown in FIG. 1 , the road image data obtained by the camera 101 and the camera position information collected by the positioning device 102 are input to the pavement type determination device 11 . As shown in FIG. 4, the pavement type determination device 11 includes an input unit 111, an input information storage unit 112, an image processing unit 113, a pavement type determination unit 114, a determination result storage unit 115, an editing/complementation processing unit 116, a determination A result output unit 117 , a display control unit 118 and a display unit 119 are provided.

The input unit 111 inputs road image data and camera position information, and associates the road image data with the camera position information.
The input information storage unit 112 records road image data and camera position information.
The image processing unit 113 performs image processing while recording the road image data in the input information storage unit 112 to extract the road surface portion of the road image.

The pavement type determination unit 114 determines the pavement type (here, porous asphalt pavement/dense-grained asphalt pavement/other pavement (semi-flexible pavement, concrete pavement, etc.) from the road surface portion of the road image obtained by the image processing unit 113. 3 types) is automatically determined.
In the judgment process, a neural network using AI (Artificial Intelligence) technology, machine learning, image processing, etc., pavement type images ("dense-grained asphalt pavement", "porous asphalt pavement") given in advance as training data ” and “Other pavement” images) , the pavement type of the image recognized as closest to the input road image is determined and output.

FIG. 5 is a diagram for explaining a teaching data creation method for AI applied to the pavement type determination unit 114. As shown in FIG.
The AI training data that automatically distinguishes the pavement type is processed from one original image (1920 x 1080 pixels) into multiple image data (768 x 768 pixels, 1080 x 1080 pixels, 96 x 96 pixels, etc.) and used. . Also, the size processing method can be freely selected, such as deleting both ends as shown in FIG. 5A or cutting out the pavement portion as shown in FIG. 5B.

The AI training data that automatically distinguishes pavement types is classified by attributes such as the weather (clear, cloudy, rain), the presence or absence of shadows, the presence or absence of cracks , and shape. Considering the attribute, an image suitable for the conditions to be applied is adopted.
When generating a learning model, all of the teacher data may be used, or one of the teacher data may be selected to generate one learning model. Alternatively, a plurality of learning models may be generated for each image characteristic, and the pavement type of one piece of input data may be determined by combining them.

It is desirable to provide versatility so that pavement type determination can be handled in environments other than the same environment (equipment, equipment installation position, shooting direction, etc.) as the training data that generated the learning model. For example, it is preferable to use a learning model generated from moving images captured by a video camera to determine images captured by an action camera or a drive recorder. In addition, it is preferable that even images captured by a video camera can be determined using a different model. However, in these cases, it is necessary to generate training data that takes into account conditions that do not depend on the equipment (eg, pavement only).

When classifying porous asphalt pavement/dense-grained asphalt pavement/other pavement, a learning model that distinguishes asphalt pavement and other pavement by considering differences in color and texture is used. In addition, after a learning model that distinguishes between asphalt pavement and other pavements based on information such as color is used, a learning model that distinguishes between asphalt pavements by taking into consideration differences in texture is used to make step-by-step judgments. may

On roads, the same pavement type tends to continue for a certain amount of section (hereinafter referred to as section A), so section A is judged at regular intervals, and the judgment results are integrated to determine the final pavement of section A. The type may be output.
The determination result storage unit 115 records the determination result of the pavement type together with the image to be determined.

The editing/complementing processing unit 116 edits/complements the information recorded in the determination result storage unit 115 visually or in cooperation with another system (road ledger system, etc.). Also, the registered information in the road management database is collated, and if there is different information or missing information, the registered information in the road management database is edited and supplemented.

The determination result output unit 117 reads out the determination results recorded in the determination result storage unit 115 sequentially or upon request, converts them into, for example, a text file and outputs them.
The display control unit 118 reads the road image data of the designated position and the pavement type determination result from the processing result storage unit 115 and converts them into a display image.

The display unit 119 displays the road image data converted by the display control unit 118 and the display image of the pavement type determination result. The screen display data is output externally upon request.
FIG. 6 is a flow chart showing the flow of processing of the pavement type determination device 11 shown in FIG.
That is , the pavement type determination device 11 determines whether the road image data and the camera position information have been input (step S11).
If there is an input , the camera position information is associated with the road image data (step S12), and then the road image data is recorded in the input information storage unit 112 (step S13).
Subsequently , the road surface portion is extracted from the recorded road image data (step S14). The extracted road surface portion is compared and collated with the pavement type image, which is pavement type training data, by AI to automatically determine whether it is porous asphalt pavement, dense-grained asphalt pavement, or other pavement (step S15). The determination result is recorded in the determination result storage unit 115 together with the image to be determined (step S16).

When there is an output request for the recorded pavement type determination result (step S17), the determination result is read from the storage unit 115, converted into, for example, a text file and output (step S18). Also, when there is a display request (step S19), the road image of the specified position and the pavement type determination result are displayed on the display unit 119 (step S1A).

Regarding the record information of the judgment result, the judgment result is monitored by the output request and the display request, and if editing/complementing is necessary (step S1B), it is edited/complemented visually or in cooperation with another system (step S1C). ).
Fig. 7 shows examples of pavement type determination results. Fig. 7 (a) shows porous asphalt pavement, Fig. 7 (b) shows dense-grained asphalt pavement, and Fig. 7 (c) shows other pavement judgment results. showing.

FIG. 8 is a block diagram showing a specific configuration of the pavement deterioration determining device 12 used in the road maintenance management system.
The type determination result file of the pavement type determination device 11 is sent to the pavement deterioration determination device 12 together with the road image data and the camera position information.
As shown in FIG. 8, the pavement deterioration determination device 12 includes an input unit 121, an input information storage unit 122, an image processing unit 123, a pavement deterioration determination unit 124, a determination result storage unit 125, a determination result output unit 126, and a display control unit. A unit 127 and a display unit 128 are provided.

The input unit 121 inputs a type determination result file determined by the pavement type determination device 11 together with road image data and camera position information, and associates the road image data with the camera position information and the pavement type determination result.
The input information storage unit 122 records road image data in which the camera position information and the pavement type determination result are associated with each other.

The image processing unit 123 performs image processing while recording the road image data in the input information storage unit 122, extracts the road surface portion of the road image, and analyzes the locations of cracks and peeling.
The pavement deterioration determination unit 124 includes a determination unit for each pavement type, that is, a drainage deterioration determination unit 124A, a dense grain system deterioration determination unit 124B, and an other deterioration determination unit 124C. Each determination unit 124A, 124B, 124C automatically determines the degree of pavement deterioration of the road surface portion of the road image obtained by the image processing unit 123 for each pavement type .
This pavement deterioration determination process also uses a neural network using AI technology, machine learning, image processing, etc., based on the deterioration image for each pavement type given in advance as teacher data, and recognizes the input road image as the closest. The degree of pavement deterioration in the image is determined and output.

Specifically, deterioration determination of porous asphalt pavement, dense-grained asphalt pavement, and other pavements is performed by, for example, detecting cracks in the pavement based on the brightness information of the road surface, and photographing the shape data of the detected cracks with the camera 101. A crack superimposed image (image with cracks in the pavement) is superimposed on the road image, and the crack rate, which indicates the ratio of the crack area to a predetermined area, is automatically calculated. It is classified into three levels (large/medium/small). Alternatively, the degree of pavement deterioration may be determined automatically from the road surface conditions (aggregate scattering, subsidence, potholes, etc.).

The determination result storage unit 125 records the image data to be determined, the crack shape data, and the deterioration determination result for each pavement type.
The determination result output unit 126 reads out the determination results recorded in the determination result storage unit 125 sequentially or upon request, converts them into, for example, a text file and outputs them.

The display control unit 127 reads out the road image data, the crack shape data, and the pavement type determination result at the designated position from the processing result storage unit 125, superimposes the crack shape image on the road image, and displays a display image including the determination result. Convert to
The display unit 128 displays the road image showing the cracks converted by the display control unit 127 and the display image of the pavement type determination result. The screen display data is output externally upon request.

FIG. 9 is a flow chart showing the flow of processing of the pavement deterioration determination device 12 shown in FIG. That is, the pavement deterioration determination device 12 determines whether the road image data, the camera position information, and the type determination result file are input (step S21). When there is an input, the road image data is associated with the camera position information and pavement type determination result (step S22), and then the road image data is recorded in the input information storage unit 122 (step S23).

Subsequently, the road surface portion is extracted from the recorded road image data (step S24), and sorted into porous asphalt pavement/dense-grade asphalt pavement/other pavement types (step S25).
A crack is detected from the road surface portion of the road image and its shape is measured (step S26).
Then , the crack rate is calculated (step S27), and the degree of deterioration for each pavement type is automatically judged, for example, in three stages (step S28).
Then , the determination result is recorded in the determination result storage unit 125 together with the image to be determined (step S29).

When there is an output request for the recorded pavement deterioration determination result (step S2A), the determination result is read from the storage unit 125, converted into a text file, for example, and output (step S2B). Also, when there is a display request (step S2C), the road image at the designated position, the shape of the crack, and the pavement deterioration determination result are displayed on the display unit 128 (step S2D).

FIG. 10 shows an example of pavement deterioration determination results. FIG. 10(a) shows the deterioration determination result of porous asphalt pavement, and FIG. 10(b) shows the deterioration determination result of dense-grained asphalt pavement.
FIG. 11 is a block diagram showing a specific configuration of the repair priority determination device 13 used in the road maintenance management system.

The determination result of the pavement deterioration determination device 12 is sent to the repair priority determination device 13 together with the road image data, the camera position information, the type determination result file, and the deterioration determination result file.
As shown in FIG. 11, the repair priority determination device 13 includes an input unit 131, an input information storage unit 132, a section data registration unit 133, a deterioration prediction formula creation unit 134, a priority evaluation unit 135, and a repair plan creation unit. 136, a processing result storage unit 137, a processing result output unit 138, a display control unit 139, and a display unit 13A.

In addition to the road image data, camera position information, type determination result file, deterioration determination result file, the input unit 131 also inputs information on section registration data (repair history, traffic volume, route type, usage, etc.), Associate each data.
The section-by-section data registration section 133 divides the input information associated by the input section 131 into sections and records them in the input information storage section 132 .

The deterioration prediction formula creation unit 134 creates a degradation prediction formula using regression analysis based on the type determination result file, degradation determination result file, and section registration data recorded in the input information recording unit 132 .
The priority evaluation unit 135 performs priority evaluation of countermeasures based on the type determination result file, the deterioration determination result file, the section registration data, and the prediction result by the deterioration prediction formula.

The repair plan creation unit 136 creates a repair plan based on the type determination result file, the deterioration determination result file, the section registration data, the prediction result by the deterioration prediction formula, and the priority evaluation result.
The processing result storage unit 137 records the deterioration prediction formula creation result, the priority evaluation result, and the repair plan creation result.

The determination result output unit 138 reads the processing result recorded in the processing result storage unit 137 in response to a request, and outputs it as a text file, for example.
In response to a request, the display control unit reads the processing result recorded in the processing result storage unit 137 and converts it into a data format that can be displayed on the screen. The display section 13A displays an image generated by the display control section 139. FIG.

FIG. 12 is a flow chart showing the flow of processing of the repair priority determination device 13 shown in FIG.
That is, the repair priority determination device 13 determines input of road image data, camera position information, type determination result file, deterioration determination result file, and section registration data (step S31).

Here, if there is an input, the input information is associated (step S32), the input information is divided into sections (step S33), and recorded in the input information storage unit 132 (step S34).
Subsequently, based on the recorded type determination result file, deterioration determination result file, and section registration data, a deterioration prediction formula is created using regression analysis (step S35).

Based on the type determination result file, deterioration determination result file, section registration data, and prediction results from the deterioration prediction formula created in this way, priority evaluation of measures to determine sections where repairs should be given priority. (step S36). Based on the type determination result file, the deterioration determination result file, the section registration data, the prediction result by the deterioration prediction formula, and the priority evaluation result, a repair plan is created (step S37) and recorded in the processing result storage unit 137 ( step S38).

When there is an output request for the recorded processing result (step S39), the processing result is read from the storage unit 137, converted into a text file, for example, and output (step S3A). Also, when there is a display request (step S3B), the road image at the specified position, the shape of the crack, the pavement deterioration determination result, and the repair priority information are displayed on the display unit 13A (step S3C).

That is, in the repair priority determination process, by registering data on section characteristics for each fixed section of the evaluation target route, a list (long list) can be created. Here, the registration data for each section includes repair history, traffic volume, route type, usage (bus routes, emergency transportation roads, etc.), and the like.

Create a deterioration prediction formula for predicting the future degree of deterioration according to the number of years elapsed for each pavement type by performing regression analysis on the relationship between the number of years since repair and the degree of deterioration for data for each fixed section. be able to. Therefore, the degree of deterioration determined/predicted based on the deterioration prediction formula and the priority of countermeasures according to the importance of the section are evaluated. The importance of a section is evaluated by using registered data for each section, selecting items to be evaluated, and weighting the items. Target items and weights between items should be arbitrarily set.

A long-term repair plan will be created based on the current degree of deterioration (determined by survey) and the future degree of deterioration (predicted by a deterioration prediction formula) in a certain section, and the results of priority evaluation for each section. In addition, the standards for repair (degree of deterioration), the repair method and cost according to the degree of deterioration, and budget constraints for each fiscal year are data to be registered at the time of planning.

FIG. 13 shows a map display example of the determination result of the repair priority determination process.
In this display example, a GIS (Geographic Information System) is used to display a circle mark for each fixed section on the specified road on the map, and the color of the circle mark is red based on the pavement type judgment result. (drainage), yellow (dense grain type), blue (others), etc. In this example, when an arbitrary ○ mark is designated (clicked), a pop-up road image of that point is displayed. Although details are not shown, it is effective if the degree of deterioration and repair priority are also displayed.

FIG. 14 shows a display example of a list (form) of determination results of the repair priority determination process. In this example, the repair priority is evaluated based on the type determination result, deterioration determination result (crack rate), and section registration data (traffic volume, route type, bus route) for each fixed section of the evaluation target route. The importance of the section is evaluated based on the section registration data, and the priority of repair is determined based on the deterioration judgment result and the section importance.

As described above, according to the road maintenance management system according to the present embodiment, it is possible to easily determine the type of pavement from the road image and the camera position information. From this determination result, it is possible to determine the degree of deterioration for each type of pavement, and even determine the priority of repair.

It should be noted that the present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying constituent elements without departing from the scope of the present invention at the implementation stage. Further, various inventions can be formed by appropriate combinations of the plurality of constituent elements disclosed in the above embodiments. For example, some components may be omitted from all components shown in the embodiments. Furthermore, components across different embodiments may be combined as appropriate.

1... road maintenance system,
10... Survey vehicle, 101... Camera, 102... Positioning device,
11... pavement type determination device, 111... input unit, 112... input information storage unit, 113... image processing unit, 114... pavement type determination unit, 115... determination result storage unit, 116... edit/complement processing unit, 117... determination result output unit, 118... display control unit, 119... display unit,
12 Pavement deterioration determination device 121 Input unit 122 Input information storage unit 123 Image processing unit 124 Pavement deterioration determination unit 125 Determination result storage unit 126 Determination result output unit 127 Display control part, 128 ... display part,
13... repair priority determination device, 131... input unit, 132... input information storage unit, 133... section data registration unit, 134... deterioration prediction formula creation unit, 135... priority evaluation unit, 136... repair plan creation unit, 137... Processing result storage unit, 138... Processing result output unit, 139... Display control unit, 13A... Display unit.

Claims (12)

An input unit for inputting a road image obtained by photographing a road and photographing position information of the road image , an extracting unit for extracting a road surface portion from the road image input to the input unit , and an extracting unit provided in advance for each pavement type. a pavement type determination unit that determines the pavement type of the road surface portion of the road image using a plurality of learning models generated by learning the pavement type image of the road surface portion with teacher data classified by attribute unit; a pavement type determination device having
An image analysis unit that detects cracks or peeling of the pavement in the road surface portion of the road image whose pavement type is determined by the pavement type determination device and analyzes the deterioration state of the pavement , and the image analysis unit. a pavement deterioration determination unit that determines the degree of pavement deterioration for each pavement type based on the detection result of the crack or peeling location of the pavement and the analysis result of the deterioration state of the pavement; and ,
Data registration for registering the road image, the shooting position information, the pavement type determination result, the pavement deterioration determination result, and input information of section registration data including management information registered for each section of the road in association with each other. and a deterioration prediction formula creation unit that creates a degradation prediction formula that predicts the degree of future deterioration using regression analysis based on the pavement type determination result, the pavement degradation determination result, and the section registration data. , a priority determination unit that determines the priority of repair for each section of the road based on the determination result of the pavement type, the determination result of the pavement deterioration, the section registration data, and the prediction result by the deterioration prediction formula; , a repair priority determination device having
A road maintenance system comprising : an input unit for inputting a road image obtained by photographing a road and photographing position information of the road image;
an extraction unit for extracting a road surface portion from the road image input to the input unit;
Using a plurality of learning models generated by learning a pavement type image of a road surface portion, which is given in advance for each pavement type, with teacher data classified by attribute unit including weather and presence or absence of shadow, the road a determination unit that determines the pavement type of the road surface portion of the image ;
A pavement type determination device comprising . an input unit for inputting a road image obtained by photographing a road and photographing position information of the road image;
an extraction unit for extracting a road surface portion from the road image input to the input unit;
Using a plurality of learning models generated by learning a pavement type image of a road surface portion, which is given in advance for each pavement type, with teacher data classified by attribute unit, the pavement type is determined for each fixed distance for the section of the road. A determination unit that determines the final pavement type of the section by integrating the determination results for each distance ,
A pavement type determination device comprising . The determination unit determines whether the pavement type of the road surface portion of the road image is porous asphalt pavement, dense-grained asphalt pavement, or other pavement .
The pavement type determination device according to claim 2 or 3 . Road image obtained by photographing the road, photographing position information of the road image, pavement type determination result for the road surface portion of the road image, pavement deterioration determination result for each pavement type for the road surface portion of the road image, section unit of the road A data registration unit that registers in association with input information of section registration data including management information registered in
A deterioration prediction formula creation unit that creates a deterioration prediction formula for predicting the degree of future deterioration using regression analysis based on the pavement type determination result, the pavement deterioration determination result, and the section registration data;
A priority determination unit that determines the priority of repair for each section of the road based on the determination result of the pavement type, the determination result of the pavement deterioration, the section registration data, and the prediction result by the deterioration prediction formula ;
A repair priority determination device comprising : A repair plan for creating a repair plan for the road based on the pavement type determination result, the pavement deterioration determination result, the section registration data, the prediction result by the deterioration prediction formula, and the repair priority determination result. the creation department ,
The repair priority determination device according to claim 5 , further comprising. A pavement type determination device inputs a road image of a road and the shooting position information of the road image,
The pavement type determination device extracts a road surface portion from the input road image,
The pavement type determination device uses a plurality of learning models generated by learning a pavement type image of a road surface portion, which is given for each pavement type in advance, with teacher data classified by attribute unit, to determine the road surface of the road image. Judge the pavement type of the part,
A pavement deterioration determination device analyzes the deterioration state of the pavement by detecting cracks or peeling of the pavement in the road surface portion of the road image for which the pavement type has been determined,
The pavement deterioration determination device determines the degree of pavement deterioration for each pavement type based on the road image, the imaging position information, the detection result of cracks or peeling of the pavement, and the analysis result of the deterioration state of the pavement. judge ,
A repair priority determination device uses regression analysis based on section registration data including the determination result of the pavement type, the determination result of the pavement deterioration, and the management information registered for each section of the road to predict future deterioration. Create a deterioration prediction formula that predicts the degree of
The repair priority determination device determines the repair priority for each section of the road based on the pavement type determination result, the pavement deterioration determination result, the section registration data, and the prediction result by the deterioration prediction formula. judge ,
Road maintenance method. Inputting a road image in which a road is photographed and photographing position information of the road image from an input unit ,
An extraction unit extracts a road surface portion from the road image,
The determination unit uses a learning model that learns a pavement type image of a road surface portion , which is given in advance for each pavement type, as teacher data classified into attribute units including weather and the presence or absence of shadows, and determines the road surface of the road image. determine the pavement type of the part ,
Pavement type determination method. Inputting a road image in which a road is photographed and photographing position information of the road image from an input unit,
An extraction unit extracts a road surface portion from the road image,
The judging unit judges the pavement type for each fixed distance for each section of the road by using a learning model in which the pavement type image of the road surface portion given in advance for each pavement type is learned as teacher data classified by attribute unit. , integrating the determination results for each distance to determine the final pavement type of the section ,
Pavement type determination method. Determination of the pavement type,
The determination unit determines whether the pavement type of the road surface portion of the road image is porous asphalt pavement, dense-grained asphalt pavement, or other pavement .
The pavement type determination method according to claim 8 or 9 . A data registration unit provides a road image obtained by photographing a road, photographing position information of the road image, pavement type determination result for the road surface portion of the road image, pavement deterioration determination result for each pavement type for the road surface portion of the road image, registering the input information of the section registration data containing the management information registered for each section of the road in association with each other;
A degradation prediction formula creation unit creates a degradation prediction formula for predicting the degree of future degradation using regression analysis based on the pavement type determination result, the pavement degradation determination result, and the section registration data,
A determination unit determines the priority of repair for each section of the road based on the determination result of the pavement type, the determination result of the pavement deterioration, the section registration data, and the prediction result by the deterioration prediction formula .
Repair priority determination method. moreover,
The repair plan creation unit repairs the road based on the determination result of the pavement type, the determination result of the pavement deterioration, the section registration data, the prediction result by the deterioration prediction formula, and the determination result of the priority of repair. create a plan ,
The repair priority determination method according to claim 11 .

Images