CN113963285A - Road maintenance method and equipment based on 5G - Google Patents
Road maintenance method and equipment based on 5G Download PDFInfo
- Publication number
- CN113963285A CN113963285A CN202111057466.1A CN202111057466A CN113963285A CN 113963285 A CN113963285 A CN 113963285A CN 202111057466 A CN202111057466 A CN 202111057466A CN 113963285 A CN113963285 A CN 113963285A
- Authority
- CN
- China
- Prior art keywords
- image
- road
- maintenance
- recognized
- disease
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 148
- 238000000034 method Methods 0.000 title claims abstract description 60
- 201000010099 disease Diseases 0.000 claims abstract description 124
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 124
- 230000004927 fusion Effects 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 2
- 230000008569 process Effects 0.000 description 13
- 230000005540 biological transmission Effects 0.000 description 7
- 230000007547 defect Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The application provides a road maintenance method and equipment based on 5G, which are used for determining traceability information corresponding to a road image. The road image is obtained by extracting image data sent by the user terminal through the 5G network. The tracing information includes the position where the road image is captured and the capturing time of the road image. And splicing the road images to obtain the images to be identified corresponding to the road images. And determining whether the image to be identified has a matched contrast image or not according to the tracing information. Wherein the contrast image is an image of a road segment. And under the condition that the image to be recognized has a matched contrast image, inputting the image to be recognized into a preset first image recognition model for recognition so as to determine a road disease label in the image to be recognized. And generating road maintenance information based on the road disease label, the tracing information and the comparison image. And sending the road maintenance information to a corresponding maintenance management terminal through a 5G network.
Description
Technical Field
The application relates to the technical field of road maintenance, in particular to a road maintenance method and equipment based on 5G.
Background
Road traffic in China is in a high-speed development period, and the development of the road traffic not only promotes the interconnection and intercommunication among regions, but also provides effective support for the development of national economy. However, roads are damaged to various degrees depending on various factors such as air temperature, weather, and rolling of vehicles. If the road damage cannot be found in time and the road maintenance is carried out, the damage degree of the road is further worsened, and the service life of the road is shortened.
Roads which are not maintained in time not only increase maintenance cost in the subsequent maintenance process, but also influence the normal traffic condition and safety problem. At present, road maintenance mainly depends on manual maintenance, and maintenance personnel drive maintenance vehicles to patrol and check on each road, so that time and energy are consumed, and normal traffic order is also hindered.
Disclosure of Invention
The embodiment of the application provides a road maintenance method and equipment based on 5G, which are used for saving labor and guaranteeing road traffic safety in road maintenance work.
In one aspect, the present application provides a road maintenance method based on 5G, including:
and determining the tracing information corresponding to the road image. The road image is obtained by extracting image data sent by the user terminal through the 5G network. The tracing information includes the position where the road image is captured and the capturing time of the road image. And splicing the road images to obtain the images to be identified corresponding to the road images. And determining whether the image to be identified has a matched contrast image or not according to the tracing information. Wherein the contrast image is an image of a road segment. And under the condition that the image to be recognized has a matched contrast image, inputting the image to be recognized into a preset first image recognition model for recognition so as to determine a road disease label in the image to be recognized. Wherein, the road disease label at least comprises one or more of the following items: cracks, web breaks, pits, ruts, subsidence. And generating road maintenance information based on the road disease label, the tracing information and the comparison image. The road maintenance information comprises maintenance time and a construction mode. And sending the road maintenance information to a corresponding maintenance management terminal through a 5G network.
In an implementation manner of the application, according to the tracing information, an undetermined area which takes a position where the shooting is located as a center and takes a preset distance as a radius is determined. A number of candidate road segments within the area to be determined are determined. And acquiring historical road images of the candidate road sections. And inputting the image to be recognized into the second image recognition model to obtain the background feature of the image to be recognized. And calculating a feature comparison value of the background feature of the image to be identified and the feature of the historical road image. And taking the historical road image with the characteristic comparison value larger than the first preset threshold value as a comparison image.
In an implementation manner of the application, the maintenance operation time of the road to be maintained corresponding to the road damage label is determined according to the road damage label. And determining the road to be maintained according to the shooting position, the shooting time and the comparison image, and determining a passing frequency curve of the road section at the shooting position. And determining a time period in which the passing frequency in the passing frequency curve is less than a second preset threshold value as the pending maintenance time. And matching the maintenance time to be determined with the maintenance operation time to obtain the maintenance time. And generating road maintenance information according to the maintenance time and the construction mode corresponding to the road disease label.
In one implementation of the present application, a capture time for each image in a road image is determined. And sequentially sequencing the road images according to the shooting time of each image to generate an image sequence to be spliced. And selecting an initial splicing image in the image sequence to be spliced. And determining the image to be spliced corresponding to the initial spliced image in the image sequence to be spliced as a first spliced image according to the initial spliced image and a preset rule. And identifying an overlapping area of the initial spliced image and the first spliced image, and carrying out image fusion on the initial spliced image and the first spliced image according to a pixel value corresponding to the overlapping area and a preset weight formula. The weight formula is used for expressing the relation between the pixel point in the overlapping area and the boundary of the overlapping area. And taking the initial spliced image after the image fusion and the first spliced image as initial spliced images, and determining the next first spliced image until the frequency of determining the first spliced image is greater than or equal to a third preset threshold value so as to obtain a spliced image of the road image. And taking the spliced image as an image to be identified.
In an implementation manner of the application, the image to be recognized is input into the feature extraction module of the first image recognition model through the input layer, so as to obtain a feature image corresponding to the image to be recognized. The feature image is input into a region recommendation network module of the first image recognition model to determine a candidate region of the feature image. The candidate areas comprise road defect areas and non-road defect areas. And inputting the candidate area into an interested area module of the first image recognition model for regression processing so as to output the road disease area of the image to be recognized. And determining a road disease label in the image to be identified according to the road disease area.
In an implementation manner of the application, a road disease area of an image to be recognized is compared with a plurality of sample images for training a first image recognition model, and a road disease label corresponding to the road disease area is determined. Or the road disease area of the image to be identified is sent to the user terminals through the 5G network, so that the road disease label corresponding to the road disease area is determined according to the first feedback information of the user terminals.
In one implementation manner of the application, the disease level and the disease level reference corresponding to the road disease label are generated and sent to the user terminal according to the road disease label. And generating the road maintenance priority of the road disease label according to the disease level selected by the user terminal and/or the disease level reference picture. And sequentially sending the road disease labels to the corresponding maintenance management terminals according to the road maintenance priority, so that the road maintenance personnel corresponding to the maintenance management terminals sequentially carry out maintenance work on the road diseases.
In an implementation manner of the application, when there is no matched comparison image in the image to be recognized, prompt information is sent to a user terminal corresponding to the image to be recognized through a 5G network, so as to determine a road section corresponding to the image to be recognized based on second feedback information of the user terminal, and the image to be recognized is used as the comparison image. Wherein, the prompt message at least comprises one or more of the following items: pictures, words, sounds.
In an implementation manner of the application, after determining a road fault label in an image to be identified, a contribution degree corresponding to the road fault label is generated. And updating the contribution degree of the user terminal according to the contribution degree, and sending the exchange goods with the updated contribution degree to the user terminal.
In another aspect, the present application provides a 5G-based road maintenance device, including:
at least one processor; and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to:
and determining the tracing information corresponding to the road image. The road image is obtained by extracting image data sent by the user terminal through the 5G network. The tracing information includes the position where the road image is captured and the capturing time of the road image. And splicing the road images to obtain the images to be identified corresponding to the road images. And determining whether the image to be identified has a matched contrast image or not according to the tracing information. Wherein the contrast image is an image of a road segment. And under the condition that the image to be recognized has a matched contrast image, inputting the image to be recognized into a preset first image recognition model for recognition so as to determine a road disease label in the image to be recognized. Wherein, the road disease label at least comprises one or more of the following items: cracks, web breaks, pits, ruts, subsidence. And generating road maintenance information based on the road disease label, the tracing information and the comparison image. The road maintenance information comprises maintenance time and a construction mode. And sending the road maintenance information to a corresponding maintenance management terminal through a 5G network.
According to the method and the device, the 5G network can be used for timely acquiring the road diseases in the road maintenance work, the road maintenance reaction can be timely due to the rapid network communication technology, the road diseases are prevented from being discovered in time, and great influences are caused on traffic safety and the like. According to the method and the device, the social force is used for determining the road diseases in the road, so that the labor burden in road maintenance work is reduced, and the labor resources are saved. The social strength is utilized to assist the progress of road maintenance work, the road maintenance work efficiency can be improved, the road disease removal speed is improved, and the road repair process is accelerated.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of a road maintenance method based on 5G in the embodiment of the present application;
FIG. 2 is a schematic flow chart of a 5G-based road maintenance method in the embodiment of the present application;
FIG. 3 is a schematic flow chart of a 5G-based road maintenance method in the embodiment of the present application;
FIG. 4 is a schematic diagram of a traffic frequency curve in the road maintenance method based on 5G in the embodiment of the present application;
FIG. 5 is a schematic structural diagram of a road maintenance device based on 5G in the embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Road maintenance is the maintenance of roads. The road and the structures and facilities on the road are maintained, the service performance of the road is maintained as much as possible, the damaged part is recovered in time, the driving safety, comfort and smoothness are ensured, and the transportation cost and time are saved; correct technical measures are adopted, the engineering quality is improved, the service life of the road is prolonged, and the reconstruction time is delayed.
At present, road maintenance work mainly depends on road maintenance workers to patrol and inspect roads of various road sections through naked eyes or machines, and whether road diseases needing to be removed exist in the roads or not is checked. The inspection in this form cannot timely reflect sudden road diseases, wastes a large amount of effective working time of road maintenance workers, and wastes the input of labor cost for road maintenance departments.
Based on this, the embodiment of the application provides a road maintenance method and equipment based on 5G, which are used for reducing the human input cost of road maintenance and improving the efficiency of road maintenance.
Various embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a flowchart of a road maintenance method based on 5G according to an embodiment of the present application. As shown in fig. 1, the 5G-based road maintenance method provided in the embodiment of the present application may include steps 101-106:
The road image is obtained by extracting image data sent by the user terminal through the 5G network. The tracing information includes the position where the road image is captured and the capturing time of the road image.
In the embodiment of the application, according to the operation of the user on the user terminal, the user terminal can send the collected information containing the image data to the server through the 5G network, and the server can analyze the received information containing the image data sent by the user terminal to obtain the image data. Information containing image data includes, but is not limited to: video, pictures, photographs.
If the information sent by the user terminal to the server is a video, the server may process the video, for example, extract each frame data in the video, or intercept the video frame in the video at a preset time interval, to obtain the image data of the video.
The 5G network is a fifth generation mobile communication network, and the peak theoretical transmission speed of the 5G network can reach 20Gbp steps, and compared with the transmission speed of the 4G network in the 21 st century, the transmission speed of the 5G network is more than ten times that of the 4G network. For the user, the transmission speed does not have a significant impact on the user's usage. However, for the transmission between devices and the operation of the devices to process data, the influence of the transmission speed is very great.
According to the method and the system, the data transmission speed between the user terminal and the server is improved through the 5G network, the road image can be timely and quickly sent and received, and great help is provided for follow-up determination of road diseases and maintenance of roads.
In the process that the user terminal sends the image data to the server, the user terminal can automatically or actively provide the shooting time and the shooting position of the image data, wherein the shooting position is the position of the user terminal, but not the position of a shot object in an image acquired by the user terminal, and the positions of the shot object and the shooting position are different. The server takes the shooting time and the shooting position of the image data as the traceability information for the use of the subsequent road maintenance method.
The user terminal may be a mobile phone, a computer, an ipad, or other terminal device, which is not specifically limited in this application.
The server is only an exemplary entity for implementing the 5G-based road maintenance method, and the implementation entity is not limited to the server, which is not specifically limited in this application.
And 102, splicing the road images by the server to obtain the images to be identified corresponding to the road images.
In the embodiment of the present application, the road image may include multiple images, for example, a video is sent by the user terminal, and the server may obtain images of multiple video frames from the video as the road image. For another example, the user terminal transmits images taken at a plurality of different angles, and the server takes the images as road images.
In the present application, the multiple images in the road image may be images including a common area captured, for example, images of video frames, and since the video is consecutive images, the common area (overlapping area) exists between the images of the frames. The method and the device can splice a plurality of images to obtain the image to be recognized with a larger view field.
Due to the limited shooting range of the lens, the user terminal may shoot a video or a plurality of images to completely represent the images of the road diseases. Therefore, the image to be identified which is complete, has a wide view angle range and contains the road diseases can be obtained by splicing the multiple images. If the splicing processing is not performed, the information (road disease image) contained in a single image is limited, which may cause the road disease identified by the server subsequently to have one-sidedness. If the server identifies the information contained in the single image in sequence, the calculation amount of the server is increased, and incoherent results may exist among the identification results of the server on each image. Through the splicing processing, the calculation amount of the server can be saved, and the accuracy of identifying the road image by the server is improved.
In this embodiment of the application, the server may splice the road images by executing the following method, as shown in fig. 2, specifically including the following steps:
in step 201, the server determines the shooting time of each image in the road image.
The server can determine the shooting time of each image in the road image according to the tracing information of the road image. Taking the road image obtained from the video as an example, according to the shooting time of the video and the time length of the video, the images can be sequentially ordered in time, and then the shooting time of each image is obtained. If the video shooting time is A time and the time length is B time, the server obtains x images from the video according to a preset time interval tThe time of image acquisition, the shooting time of the ith image in the x images is as follows:
wherein, the video shooting time is the shooting starting time.
It is known to those skilled in the art that the video capturing time a may also be the video capturing termination time, and when the video capturing time is the video capturing termination time, the capturing time of the image is calculated, and the capturing time of each image can be obtained by simply adjusting the above formula.
Step 202, the server sequences the road images in sequence according to the shooting time of each image so as to generate an image sequence to be spliced.
And sequencing the images in sequence according to the shooting time of each image obtained in the step 201 to obtain an image sequence to be spliced.
In the embodiment of the application, the initial stitched image in the image sequence to be stitched is an image at a middle position in the image sequence to be stitched. For example, the image sequence to be stitched is [ a, b, c, d, e ], and the server selects c as the initial stitched image. When two images exist in the middle position, such as [ a, b, c, d, e, f ], the server can select one image as an initial splicing image.
And 204, the server determines an image to be stitched corresponding to the initial stitched image in the image sequence to be stitched as a first stitched image according to the initial stitched image and a preset rule.
After obtaining the initial splicing image, the server determines a plurality of images to be spliced on any side of the initial splicing image, such as an image sequence to be spliced of [ a, b, c, d, e ], according to a preset rule, wherein c is the initial splicing image, and then determines d, e or a, b; and when the determined multiple images to be spliced are d and e, determining d adjacent to the initial spliced image c as a first spliced image. Similarly, when the determined multiple images to be stitched are a and b, taking b as the first stitched image.
The weight formula is used for expressing the relation between the pixel point in the overlapping area and the boundary of the overlapping area.
The server may identify an overlap region of the initial stitched image and the first stitched image, which may be determined by an image registration technique. After the overlapping area is determined, according to a preset weight formula, according to the pixel value in the overlapping area:
wherein the content of the first and second substances,
is a weight, dmaxIs the maximum value of the boundary of the overlap region, dminAnd d is the position of the pixel point in the overlapping area. Taking a plane coordinate system as an example, the minimum value of the boundary of the overlapping region is 3, the maximum value of the boundary of the overlapping region is 6, and the position of the pixel point is 4.
Determining the weight of each pixel point in the overlapping area according to the weight formula, and then determining the pixel value of each pixel point in the spliced overlapping area according to the following formula:
wherein Q (x, y) is the pixel value of the pixel point at the (x, y) position in the overlapping region after image fusion, and Q1(x, y) is the pixel value of the pixel point at the position of the initial splicing image (x, y) before image fusion, Q2(x, y) is the pixel value of the pixel point at the (x, y) position of the first stitched image before image fusion, R1For regions belonging to the initially stitched image，R2To belong to the overlapping region of the initial stitched image and the first stitched image, R3Is the region belonging to the first stitched image.
The server performs image fusion on the initial splicing image and the first splicing image through the formula, so that the effect of gradually transitioning from the initial splicing image to the first splicing image can be realized, and the splicing effect is good.
And step 206, the server takes the initial spliced image after the image fusion and the first spliced image as initial spliced images, and determines the next first spliced image until the frequency of determining the first spliced image is greater than or equal to a third preset threshold value, so as to obtain a spliced image of the road image.
And after the server finishes the image fusion of the initial splicing image and the first splicing image, the server updates the initial splicing image and the first splicing image which are subjected to the fusion into a new initial splicing image, and determines the first splicing image again. In the process of determining the first stitched image again, the server may determine a new first stitched image according to the rule of determining the first stitched image last time.
For example, the image d to be stitched is selected last in step 204, and this time e can be taken as the first stitched image. In the application, the server can firstly complete the splicing of the image to be spliced on one side of the initial spliced image and the initial spliced image, and then splice the image to be spliced on the other side of the initial spliced image and the initial spliced image. Through two image splicing in turn of the initial splicing, the overlapping rate can be improved, such as the overlapping area exists between the images b and d to be spliced, when the image to be spliced on one side d is spliced, the overlapping area between the images b and the initial splicing can be improved, and the overlapping rate is increased.
Until it is determined that the number of times of the first stitched image is greater than or equal to a third preset threshold, the third preset threshold may be set to be the number of the road images minus 1, so as to meet the requirement that all the road images are subjected to image stitching, or the third preset threshold is set to be an optimal threshold according to the actual stitching use condition, which is not specifically limited herein.
And step 207, the server takes the spliced image as an image to be identified.
In the process of determining the image to be identified, the technical scheme can reduce the splicing error between the spliced images in the image splicing process. The intermediate image is used as the initial splicing image, so that the problem of the distance between the images considered in image splicing is reduced, and the calculation complexity of the server in the image splicing process can be reduced; by sequentially splicing the images on the two sides of the initial spliced image, the overlapping rate during splicing can be improved, and a better image splicing effect is realized.
And 103, the server determines whether the image to be identified has a matched contrast image according to the tracing information.
Wherein the contrast image is an image of a road segment.
After the server obtains the image to be recognized, whether a matched contrast image exists in the image to be recognized can be determined according to the traceability information of the road image spliced with the image to be recognized. In the present application, whether there is a matching comparison image in the image to be recognized may be determined by performing the following method, specifically as follows:
firstly, the server determines an undetermined area which takes the position where the camera is located as the center and takes the preset distance as the radius according to the source tracing information.
And the server determines the shooting position of each road image corresponding to the image to be identified according to the tracing information, and determines an area to be determined by taking the shooting position as a center and a preset distance as a radius. The undetermined area contains a plurality of roads or a single road.
The road images corresponding to the images to be recognized come from different user terminals, and the geometric centers of the positions where the road images are shot can be determined according to the positions where the road images are shot corresponding to the road images. And the geometric center is taken as the center of a circle of the region to be determined.
The server then determines a number of candidate road segments within the area to be determined.
And the server determines the road sections contained in the undetermined area, for example, the road sections containing a plurality of transverse buses in the undetermined area are taken as candidate road sections by the server.
Next, the server acquires a history road image of each candidate link.
The server may acquire a pre-stored historical road image, or acquire the historical road image through the internet, and the acquisition mode of the historical road image is not particularly limited in the present application.
In the embodiment of the application, the shooting time of the historical road image is close to the shooting time of the road image for splicing the images to be recognized, and as can be known by a person skilled in the art, the closer the shooting time of the historical road image is to the shooting time of the road image for splicing the images to be recognized, the higher the data accuracy is.
And then, the server inputs the image to be recognized into the second image recognition model to obtain the background feature of the image to be recognized.
In this embodiment of the application, the server may train a second image recognition model in advance before determining the comparison image, where the second image recognition model trains by using each road image as sample data, and outputs an image background feature of each road image.
After the server inputs the image to be recognized into the second image recognition model, the second image recognition model processes the image to be recognized, and the background feature of the image to be recognized can be input.
And secondly, the server calculates a feature comparison value of the background feature of the image to be identified and the feature of the historical road image.
The server can generate a feature vector of the background feature of the image to be recognized and a feature vector of the feature of each historical road image, and the server can calculate the similarity of the background feature and the feature vectors of the features of the historical road images to serve as a feature comparison value.
And finally, the server takes the historical road image with the characteristic comparison value larger than the first preset threshold value as a comparison image.
The server may compare the calculated feature comparison values with a first preset threshold, and determine a historical road image corresponding to the feature comparison value larger than the first preset threshold as a comparison image. If the server determines that two or more feature comparison values satisfy the condition that the feature comparison values are larger than the first preset threshold, the server may further determine, as the comparison image, the historical road image corresponding to the feature comparison value with the largest value among the plurality of feature comparison values.
In the embodiment of the application, the road section corresponding to the image to be recognized can be more accurately and efficiently determined by determining the comparison image, and the road maintenance personnel are not required to further confirm the position of the road section to which the image to be recognized belongs. By the scheme, the efficiency of road maintenance is further improved, and the consumption of human resources is saved.
And step 104, under the condition that the image to be recognized has the matched contrast image, the server inputs the image to be recognized into a preset first image recognition model for recognition so as to determine a road disease label in the image to be recognized.
Wherein, the road disease label at least comprises one or more of the following items: cracks, web breaks, pits, ruts, subsidence.
According to the embodiment of the application, the road disease label of the image to be identified can be identified through the first image identification model. The method comprises the following specific steps:
firstly, the server inputs the image to be recognized into the feature extraction module of the first image recognition model through the input layer so as to obtain the feature image corresponding to the image to be recognized.
The server can identify an image containing a road disease in the image to be identified through the first image identification model, such as: the images of cracks, network cracks, pits, ruts and subsidence are extracted, and the characteristic extraction module can adopt a VGG (variable gradient G) model and can also adopt a model of a residual error network. In the embodiment of the application, more identifiable disease tags may exist in the road disease tag along with the evolution of the road.
Then, the server inputs the feature image into the region recommendation network module of the first image recognition model to determine a candidate region of the feature image.
The candidate areas comprise road defect areas and non-road defect areas.
The feature image is processed through a Region recommendation Network (RPN), the RPN takes the feature image after feature extraction as input, each point in each feature image is processed, and then a candidate Region is obtained. The candidate areas may include road damaged areas, non-road damaged areas.
Secondly, the server inputs the candidate area into an interested area module of the first image recognition model for regression processing so as to output a road disease area of the image to be recognized.
And the server performs pooling treatment on the obtained candidate areas by using the region-of-interest model, and then performs classification and accurate regression of the boundary positions of the candidate areas after the pooling treatment, so as to determine the road disease areas of the images to be identified.
And finally, the server determines a road disease label in the image to be identified according to the road disease area.
The server compares the road disease area of the image to be recognized with a plurality of sample images for training the first image recognition model, and determines a road disease label corresponding to the road disease area. Alternatively, the first and second electrodes may be,
the server sends the road disease area of the image to be identified to the user terminals through the 5G network, and determines road disease labels corresponding to the road disease area according to the first feedback information of the user terminals.
In the actual use process, the way of determining the road damage label corresponding to the road damage area is various, the two reference ways are provided in the application, and the application does not specifically limit other embodiments on the premise that the road damage label corresponding to the road damage area can be determined.
In addition, the server can send the road disease area of the image to be identified to a plurality of user terminals through a 5G network, so that the user can confirm whether the identified road disease area is correct and whether the determined road disease label is accurate. And based on the operation of the user on the user terminal, inputting the road disease area determined by the user and the road disease label into the first image recognition model, and retraining the first image recognition model. And the road disease label sent by the user terminal is used as the road disease label of the image to be identified.
By the scheme, the road disease image included in the image to be identified can be accurately identified, and reliable data support is provided for subsequent road maintenance. In addition, the first image recognition model is a neural network model, and R-CNN can be adopted, and other neural network models can also be adopted. The second image recognition model is also a neural network model, which may be the same as or different from the first image recognition model, and the models selected by the two do not directly influence the scheme.
According to the scheme, the road disease label is determined under the condition that the matched contrast image exists in the image to be identified, and the road disease label can be determined through the following method under the condition that the matched contrast image does not exist in the image to be identified.
Specifically, when the image to be recognized does not have a matching comparison image, the server first sends prompt information to the user terminal corresponding to the image to be recognized through the 5G network, so as to determine the road section corresponding to the image to be recognized based on the second feedback information of the user terminal, and uses the image to be recognized as the comparison image.
Wherein, the prompt message at least comprises one or more of the following items: pictures, words, sounds.
The server may send prompt information to the user terminal that sends the image to be recognized, such as: and the user can operate the user terminal on the road section corresponding to the image and respond to the prompt information sent by the server. The second feedback information may be text, picture, sound, such as text: road a, lane B, etc.
Then, the image to be recognized of the determined road section is input into the first image recognition model so as to determine the road disease label of the image to be recognized.
After the road section of the image to be identified sent by the user terminal is determined, the user terminal can be given corresponding rewards, and the user is encouraged to shoot the road image containing the road diseases. The method comprises the following specific steps:
and after determining the road disease label in the image to be identified, the server generates the contribution degree corresponding to the road disease label.
And the server updates the contribution degree of the acquisition terminal according to the contribution degree and sends the exchange goods with the updated contribution degree to the acquisition terminal.
In the application, the server may record the contribution degree of the user terminal, for example, the user terminal sends the image to be identified with the road fault label a to the server, and after the server confirms the image, the recorded contribution degree of the user terminal may be updated. The updating method may be to increase, for example, if the current contribution degree of the user terminal a is 90, after the server confirms the image to be identified with the road fault label as a, the current contribution degree is updated to 95.
In another embodiment of the present application, the server determines that the user terminal has transmitted a road image that does not include a road disease more than a certain number of times, and the server may update the contribution degree of the user terminal in a reduced manner.
In the embodiment of the application, after the road disease label corresponding to the image to be identified is determined, the severity of the road disease can be further determined, and a corresponding response can be made to the disease in time.
Specifically, the server generates and transmits a damage level and a damage level reference map corresponding to the road damage label to the user terminal according to the road damage label.
Then, the server generates a road maintenance priority of the road damage label according to the damage level selected by the user terminal and/or a reference of the damage level.
And finally, the server sequentially sends the road disease labels to the corresponding maintenance management terminals according to the road maintenance priority, so that the road maintenance personnel corresponding to the maintenance management terminals sequentially carry out maintenance work on the road diseases.
The road maintenance priority is generated according to the disease level of the road disease, so that road maintenance personnel can be helped to select the road disease with higher severity degree for treatment, the influence of the serious road disease on traffic safety is reduced, and the serious traffic safety problem is avoided. Meanwhile, the working efficiency of the road maintenance personnel is also improved. The road disease grade is confirmed by the user terminal, so that road maintenance procedures can be saved, procedures such as in-person confirmation of road maintenance personnel and the like are not needed, and labor waste is saved to a certain extent.
And 105, the server generates road maintenance information based on the road disease label, the tracing information and the comparison image.
The road maintenance information comprises maintenance time and a construction mode.
The method for generating the road maintenance information comprises the following steps of:
After the road damage label is determined, the server may determine, according to the road damage label, the required maintenance operation time for the road maintenance worker to treat the road damage corresponding to the road damage label through the 5G network or the pre-stored operation time comparison table.
Step 302, the server determines the road to be maintained according to the shooting position, the shooting time and the comparison image, and determines a passing frequency curve of the road section at the shooting position.
According to the shooting position and the comparison image, a road section such as the section i of the road at the shooting position can be determined. After the photographing time, the traffic frequency curve of the predicted section i may be determined by the recorded data of the historical driving of the section i according to the photographing time of the road image, as shown in fig. 4, y is traffic efficiency, and x is time.
According to the passing frequency curve, the server may determine the maintenance time to be determined, as shown in fig. 4, the dotted line is the passing frequency corresponding to the second preset threshold. In fig. 4, the time corresponding to the portion of the passing frequency curve between the dotted line and the x-axis is the pending maintenance time.
And step 304, the server matches the maintenance time to be determined with the maintenance operation time to obtain the maintenance time.
The server can match the maintenance waiting time with the maintenance operation time, for example, the maintenance operation time is t1, and the maintenance waiting time has a plurality of time periods with different time lengths. For example, the undetermined maintenance time includes t2, t3 and t4, and three time periods, wherein t2< t1, t3> t1 and t4< t1, the server may use t3 as the maintenance time.
And 305, generating road maintenance information by the server according to the maintenance time and the construction mode corresponding to the road defect label.
The server can obtain a corresponding construction mode through a 5G network according to the road disease label, and generate road maintenance information according to the maintenance time and the obtained construction mode.
And 106, the server sends the road maintenance information to a corresponding maintenance management terminal through a 5G network.
The maintenance management terminal may correspond to a road maintenance department, and may also be a hand-held terminal of a patrol officer of the road maintenance department, or the like. The specific type of the maintenance management terminal is not specifically limited herein.
The method and the device receive image data sent by a user terminal through a 5G network, determine a road image, and determine the shooting position and the shooting time of the road image. And splicing the road images to obtain an image to be identified, and further determining a road disease label corresponding to the image to be identified. And generating and sending road maintenance information to a maintenance management terminal according to the road disease label. The user can participate in the road maintenance work, the workload of a road maintenance department can be saved, and the labor cost is reduced. The 5G network is used for receiving and sending data, so that information in road maintenance work can be provided more timely, and serious traffic safety accidents caused by the fact that the road maintenance information is not timely enough are avoided.
In addition, the first image recognition model is used for recognizing the road diseases, so that one-sided factors brought by artificial judgment of the diseases can be avoided, and the accuracy of recognizing the road diseases can be ensured. The road diseases are accurately identified, and the road maintenance work efficiency can be accelerated to a certain extent.
Fig. 5 is a schematic structural diagram of a 5G-based road maintenance device 500 provided in an embodiment of the present application, where the device includes:
at least one processor 501. And a memory 502 communicatively coupled to the at least one processor 501. Wherein the memory 502 stores instructions executable by the at least one processor 501, the instructions being executable by the at least one processor 501 to enable the at least one processor 501 to:
and determining the tracing information corresponding to the road image. The road image is obtained by extracting image data sent by the user terminal through the 5G network. The tracing information includes the position where the road image is captured and the capturing time of the road image. And splicing the road images to obtain the images to be identified corresponding to the road images. And determining whether the image to be identified has a matched contrast image or not according to the tracing information. Wherein the contrast image is an image of a road segment. And under the condition that the image to be recognized has a matched contrast image, inputting the image to be recognized into a preset first image recognition model for recognition so as to determine a road disease label in the image to be recognized. Wherein, the road disease label at least comprises one or more of the following items: cracks, web breaks, pits, ruts, subsidence. And generating road maintenance information based on the road disease label, the tracing information and the comparison image. The road maintenance information comprises maintenance time and a construction mode. And sending the road maintenance information to a corresponding maintenance management terminal through a 5G network.
The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The devices and the methods provided by the embodiment of the application are in one-to-one correspondence, so the devices also have beneficial technical effects similar to the corresponding methods.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A road maintenance method based on 5G is characterized by comprising the following steps:
determining tracing information corresponding to the road image; the road image is obtained by extracting image data sent by a user terminal through a 5G network; the tracing information comprises the shooting position of the road image and the shooting time of the road image;
splicing the road images to obtain images to be identified corresponding to the road images;
determining whether the image to be identified has a matched contrast image or not according to the tracing information; wherein the contrast image is an image of a road section;
under the condition that the image to be recognized has the matched comparison image, inputting the image to be recognized into a preset first image recognition model for recognition so as to determine a road disease label in the image to be recognized; wherein the road disease label at least comprises one or more of the following items: cracks, web breaks, pits, ruts, subsidence;
generating road maintenance information based on the road disease label, the traceability information and the comparison image; the road maintenance information comprises maintenance time and a construction mode;
and sending the road maintenance information to a corresponding maintenance management terminal through the 5G network.
2. The method according to claim 1, wherein determining whether the image to be recognized has a matching comparison image specifically comprises:
according to the source tracing information, determining an undetermined area which takes the position where the camera is located as the center and takes a preset distance as the radius;
determining a plurality of candidate road sections in the region to be determined;
acquiring historical road images of the candidate road sections; and
inputting the image to be recognized into a second image recognition model to obtain background features of the image to be recognized;
calculating a feature comparison value of the background feature of the image to be identified and the feature of the historical road image;
and taking the historical road image with the characteristic comparison value larger than a first preset threshold value as the comparison image.
3. The method according to claim 1, wherein generating road maintenance information based on the road disease label, the traceability information, and the comparison image specifically comprises:
determining the maintenance operation time of the road to be maintained corresponding to the road disease label according to the road disease label;
determining the road to be maintained and a traffic frequency curve of a road section at the shooting position according to the shooting position, the shooting time and the comparison image;
determining a time period of which the passing frequency is smaller than a second preset threshold in the passing frequency curve as to-be-determined maintenance time;
matching the undetermined maintenance time with the maintenance operation time to obtain maintenance time;
and generating road maintenance information according to the maintenance time and the construction mode corresponding to the road disease label.
4. The method according to claim 1, wherein the stitching processing is performed on the road image to obtain an image to be identified corresponding to the road image, and specifically includes:
determining the shooting time of each image in the road image;
sequentially sequencing the road images according to the shooting time of each image to generate an image sequence to be spliced;
selecting an initial splicing image in the image sequence to be spliced;
determining an image to be stitched corresponding to the initial stitched image in the image sequence to be stitched as a first stitched image according to the initial stitched image and a preset rule;
identifying an overlapping area of the initial spliced image and the first spliced image, and carrying out image fusion on the initial spliced image and the first spliced image according to a pixel value corresponding to the overlapping area and a preset weight formula; wherein the weight formula is used for representing the relation between the pixel point in the overlapping area and the boundary of the overlapping area;
taking the initial spliced image after image fusion and the first spliced image as the initial spliced image, and determining the next first spliced image until the frequency of determining the first spliced image is greater than or equal to a third preset threshold value, so as to obtain a spliced image of the road image;
and taking the spliced image as the image to be identified.
5. The method according to claim 1, wherein the image to be recognized is input into a preset first image recognition model for recognition, so as to determine a road disease label in the image to be recognized, specifically comprising:
inputting the image to be recognized into a feature extraction module of the first image recognition model through an input layer to obtain a feature image corresponding to the image to be recognized;
inputting the feature image into a region recommendation network module of the first image recognition model to determine a candidate region of the feature image; the candidate areas comprise road disease areas and non-road disease areas;
inputting the candidate area into an area-of-interest module of the first image identification model for regression processing so as to output the road disease area of the image to be identified;
and determining a road disease label in the image to be identified according to the road disease area.
6. The method according to claim 5, wherein determining the road damage label in the image to be identified according to the road damage area specifically comprises:
comparing the road disease area of the image to be recognized with a plurality of sample images for training the first image recognition model, and determining a road disease label corresponding to the road disease area; alternatively, the first and second electrodes may be,
and sending the road disease area of the image to be identified to a plurality of user terminals through the 5G network, so as to determine a road disease label corresponding to the road disease area according to the first feedback information of the user terminals.
7. The method of claim 6, wherein after determining the road disease label in the image to be identified, the method further comprises:
generating and sending a disease level corresponding to the road disease label and a disease level reference to the user terminal according to the road disease label;
generating a road maintenance priority of the road disease label according to the disease level selected by the user terminal and/or the disease level reference map;
and sequentially sending the road disease labels to the corresponding maintenance management terminals according to the road maintenance priority, so that the road maintenance personnel corresponding to the maintenance management terminals sequentially carry out maintenance work on the road diseases.
8. The method of claim 1, further comprising:
under the condition that the image to be recognized does not have the matched comparison image, sending prompt information to the user terminal corresponding to the image to be recognized through the 5G network so as to determine a road section corresponding to the image to be recognized based on second feedback information of the user terminal, and taking the image to be recognized as the comparison image; wherein the prompt message at least comprises one or more of the following items: pictures, words, sounds.
9. The method of claim 8, further comprising:
after determining the road disease label in the image to be identified, generating the contribution degree corresponding to the road disease label;
and updating the contribution degree of the user terminal according to the contribution degree, and sending the exchange goods of the updated contribution degree to the user terminal.
10. A road maintenance device based on 5G, characterized in that the device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to cause the at least one processor to:
determining tracing information corresponding to the road image; the road image is obtained by extracting image data sent by a user terminal through a 5G network; the tracing information comprises the shooting position of the road image and the shooting time of the road image;
splicing the road images to obtain images to be identified corresponding to the road images;
determining whether the image to be identified has a matched contrast image or not according to the tracing information; wherein the contrast image is an image of a road section;
under the condition that the image to be recognized has the matched comparison image, inputting the image to be recognized into a preset first image recognition model for recognition so as to determine a road disease label in the image to be recognized; wherein the road disease label at least comprises one or more of the following items: cracks, web breaks, pits, ruts, subsidence;
generating road maintenance information based on the road disease label, the traceability information and the comparison image; the road maintenance information comprises maintenance time and a construction mode;
and sending the road maintenance information to a corresponding maintenance management terminal through the 5G network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111057466.1A CN113963285B (en) | 2021-09-09 | 2021-09-09 | Road maintenance method and equipment based on 5G |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111057466.1A CN113963285B (en) | 2021-09-09 | 2021-09-09 | Road maintenance method and equipment based on 5G |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113963285A true CN113963285A (en) | 2022-01-21 |
| CN113963285B CN113963285B (en) | 2022-06-10 |
Family
ID=79461237
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111057466.1A Active CN113963285B (en) | 2021-09-09 | 2021-09-09 | Road maintenance method and equipment based on 5G |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113963285B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114971418A (en) * | 2022-07-14 | 2022-08-30 | 唐山陆达公路养护有限公司 | Intelligent management system of data based on road maintenance |
| CN115984273A (en) * | 2023-03-20 | 2023-04-18 | 深圳思谋信息科技有限公司 | Road disease detection method and device, computer equipment and readable storage medium |
| CN117273477A (en) * | 2023-08-16 | 2023-12-22 | 湖北交投智能检测股份有限公司 | Asphalt highway pavement maintenance decision method, device, equipment and storage medium |
| CN117575878A (en) * | 2023-11-16 | 2024-02-20 | 杭州众诚咨询监理有限公司 | Intelligent management method and device for traffic facility asset data, electronic equipment and medium |
Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002297611A (en) * | 2001-03-29 | 2002-10-11 | Toshiba Corp | Map information acquiring method, position information acquiring method, map information acquiring device, client device, map information storage device, and program |
| US20080268876A1 (en) * | 2007-04-24 | 2008-10-30 | Natasha Gelfand | Method, Device, Mobile Terminal, and Computer Program Product for a Point of Interest Based Scheme for Improving Mobile Visual Searching Functionalities |
| JP2017102672A (en) * | 2015-12-01 | 2017-06-08 | 株式会社日立ソリューションズ | Geographic position information specification system and geographic position information specification method |
| CN107085761A (en) * | 2017-03-09 | 2017-08-22 | 深圳高速工程信息有限公司 | Road maintenance methods, devices and systems |
| CN107153928A (en) * | 2017-06-28 | 2017-09-12 | 江苏智通交通科技有限公司 | Visual highway maintenance decision system |
| CN108037723A (en) * | 2018-01-24 | 2018-05-15 | 武汉浩宇天辰科技有限公司 | A kind of highway maintenance Information Management System |
| CN108734306A (en) * | 2018-04-03 | 2018-11-02 | 四川隧唐科技股份有限公司 | A kind of data processing method, device, road upkeep system and storage medium |
| CN108898085A (en) * | 2018-06-20 | 2018-11-27 | 安徽大学 | A kind of road disease intelligent detecting method based on mobile video |
| JP2019028922A (en) * | 2017-08-03 | 2019-02-21 | パナソニックＩｐマネジメント株式会社 | Information storage system, information distribution system, driving support system, and information storage method |
| US20190107400A1 (en) * | 2017-08-03 | 2019-04-11 | aiPod, Inc. | Localization of Autonomous Vehicles via Ground Image Recognition |
| CN109685124A (en) * | 2018-12-14 | 2019-04-26 | 斑马网络技术有限公司 | Road disease recognition methods neural network based and device |
| CN109716108A (en) * | 2016-12-30 | 2019-05-03 | 同济大学 | A kind of Asphalt Pavement Damage detection system based on binocular image analysis |
| US20200074413A1 (en) * | 2018-08-28 | 2020-03-05 | Kabushiki Kaisha Toshiba | Road maintenance management system, road maintenance management method, and a non-transitory recording medium |
| CN110894704A (en) * | 2019-11-15 | 2020-03-20 | 东南大学 | Unmanned aerial vehicle-based method for positioning road surface diseases |
| CN111047585A (en) * | 2019-12-25 | 2020-04-21 | 苏州奥易克斯汽车电子有限公司 | Pavement crack detection method |
| CN111062871A (en) * | 2019-12-17 | 2020-04-24 | 腾讯科技（深圳）有限公司 | Image processing method and device, computer equipment and readable storage medium |
| CN111126802A (en) * | 2019-12-10 | 2020-05-08 | 福建省高速公路集团有限公司 | Highway inspection and evaluation method and system based on artificial intelligence |
| CN111435088A (en) * | 2019-01-15 | 2020-07-21 | 北京嘀嘀无限科技发展有限公司 | Road matching method and device, electronic equipment and storage medium |
| CN112307989A (en) * | 2020-11-03 | 2021-02-02 | 广州海格通信集团股份有限公司 | Method and device for identifying road surface object, computer equipment and storage medium |
| WO2021017212A1 (en) * | 2019-07-26 | 2021-02-04 | 魔门塔(苏州)科技有限公司 | Multi-scene high-precision vehicle positioning method and apparatus, and vehicle-mounted terminal |
| CN112465166A (en) * | 2020-11-24 | 2021-03-09 | 天津市公路工程总公司 | Highway minor maintenance management system and method |
| US20210097311A1 (en) * | 2019-09-27 | 2021-04-01 | Dish Network L.L.C. | Wireless vehicular systems and methods for detecting roadway conditions |
| WO2021076573A1 (en) * | 2019-10-15 | 2021-04-22 | Roadbotics, Inc. | Systems and methods for assessing infrastructure |
| CN112800913A (en) * | 2021-01-20 | 2021-05-14 | 同济大学 | Pavement damage data space-time analysis method based on multi-source feature fusion |
| CN112861706A (en) * | 2021-02-04 | 2021-05-28 | 北京百度网讯科技有限公司 | Road state monitoring method, device, equipment and storage medium |
| CN113033840A (en) * | 2021-03-29 | 2021-06-25 | 唐山市曹妃甸区陆月柒峰科技有限责任公司 | Method and device for judging highway maintenance |
| CN113256969A (en) * | 2021-04-30 | 2021-08-13 | 济南金宇公路产业发展有限公司 | Traffic accident early warning method, device and medium for expressway |
| CN113283285A (en) * | 2021-03-19 | 2021-08-20 | 南京四维向量科技有限公司 | Method for accurately positioning address based on image recognition technology |
-
2021
- 2021-09-09 CN CN202111057466.1A patent/CN113963285B/en active Active
Patent Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002297611A (en) * | 2001-03-29 | 2002-10-11 | Toshiba Corp | Map information acquiring method, position information acquiring method, map information acquiring device, client device, map information storage device, and program |
| US20080268876A1 (en) * | 2007-04-24 | 2008-10-30 | Natasha Gelfand | Method, Device, Mobile Terminal, and Computer Program Product for a Point of Interest Based Scheme for Improving Mobile Visual Searching Functionalities |
| JP2017102672A (en) * | 2015-12-01 | 2017-06-08 | 株式会社日立ソリューションズ | Geographic position information specification system and geographic position information specification method |
| CN109716108A (en) * | 2016-12-30 | 2019-05-03 | 同济大学 | A kind of Asphalt Pavement Damage detection system based on binocular image analysis |
| CN107085761A (en) * | 2017-03-09 | 2017-08-22 | 深圳高速工程信息有限公司 | Road maintenance methods, devices and systems |
| CN107153928A (en) * | 2017-06-28 | 2017-09-12 | 江苏智通交通科技有限公司 | Visual highway maintenance decision system |
| US20190107400A1 (en) * | 2017-08-03 | 2019-04-11 | aiPod, Inc. | Localization of Autonomous Vehicles via Ground Image Recognition |
| JP2019028922A (en) * | 2017-08-03 | 2019-02-21 | パナソニックＩｐマネジメント株式会社 | Information storage system, information distribution system, driving support system, and information storage method |
| CN108037723A (en) * | 2018-01-24 | 2018-05-15 | 武汉浩宇天辰科技有限公司 | A kind of highway maintenance Information Management System |
| CN108734306A (en) * | 2018-04-03 | 2018-11-02 | 四川隧唐科技股份有限公司 | A kind of data processing method, device, road upkeep system and storage medium |
| CN108898085A (en) * | 2018-06-20 | 2018-11-27 | 安徽大学 | A kind of road disease intelligent detecting method based on mobile video |
| US20200074413A1 (en) * | 2018-08-28 | 2020-03-05 | Kabushiki Kaisha Toshiba | Road maintenance management system, road maintenance management method, and a non-transitory recording medium |
| CN109685124A (en) * | 2018-12-14 | 2019-04-26 | 斑马网络技术有限公司 | Road disease recognition methods neural network based and device |
| CN111435088A (en) * | 2019-01-15 | 2020-07-21 | 北京嘀嘀无限科技发展有限公司 | Road matching method and device, electronic equipment and storage medium |
| WO2021017212A1 (en) * | 2019-07-26 | 2021-02-04 | 魔门塔(苏州)科技有限公司 | Multi-scene high-precision vehicle positioning method and apparatus, and vehicle-mounted terminal |
| US20210097311A1 (en) * | 2019-09-27 | 2021-04-01 | Dish Network L.L.C. | Wireless vehicular systems and methods for detecting roadway conditions |
| WO2021076573A1 (en) * | 2019-10-15 | 2021-04-22 | Roadbotics, Inc. | Systems and methods for assessing infrastructure |
| CN110894704A (en) * | 2019-11-15 | 2020-03-20 | 东南大学 | Unmanned aerial vehicle-based method for positioning road surface diseases |
| CN111126802A (en) * | 2019-12-10 | 2020-05-08 | 福建省高速公路集团有限公司 | Highway inspection and evaluation method and system based on artificial intelligence |
| CN111062871A (en) * | 2019-12-17 | 2020-04-24 | 腾讯科技（深圳）有限公司 | Image processing method and device, computer equipment and readable storage medium |
| CN111047585A (en) * | 2019-12-25 | 2020-04-21 | 苏州奥易克斯汽车电子有限公司 | Pavement crack detection method |
| CN112307989A (en) * | 2020-11-03 | 2021-02-02 | 广州海格通信集团股份有限公司 | Method and device for identifying road surface object, computer equipment and storage medium |
| CN112465166A (en) * | 2020-11-24 | 2021-03-09 | 天津市公路工程总公司 | Highway minor maintenance management system and method |
| CN112800913A (en) * | 2021-01-20 | 2021-05-14 | 同济大学 | Pavement damage data space-time analysis method based on multi-source feature fusion |
| CN112861706A (en) * | 2021-02-04 | 2021-05-28 | 北京百度网讯科技有限公司 | Road state monitoring method, device, equipment and storage medium |
| CN113283285A (en) * | 2021-03-19 | 2021-08-20 | 南京四维向量科技有限公司 | Method for accurately positioning address based on image recognition technology |
| CN113033840A (en) * | 2021-03-29 | 2021-06-25 | 唐山市曹妃甸区陆月柒峰科技有限责任公司 | Method and device for judging highway maintenance |
| CN113256969A (en) * | 2021-04-30 | 2021-08-13 | 济南金宇公路产业发展有限公司 | Traffic accident early warning method, device and medium for expressway |
Non-Patent Citations (3)
| Title |
|---|
| CE ZHANG 等: "road maintenance Pavement Distress neural network", 《INTERNATIONAL JOURNAL OF TRANSPORTATION SCIENCE AND TECHNOLOGY》 * |
| 姚琳怡: "基于强化学习的高速公路项目级养护决策研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 陈曼妮: "高速公路路面养护管理信息系统设计与应用", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114971418A (en) * | 2022-07-14 | 2022-08-30 | 唐山陆达公路养护有限公司 | Intelligent management system of data based on road maintenance |
| CN115984273A (en) * | 2023-03-20 | 2023-04-18 | 深圳思谋信息科技有限公司 | Road disease detection method and device, computer equipment and readable storage medium |
| CN117273477A (en) * | 2023-08-16 | 2023-12-22 | 湖北交投智能检测股份有限公司 | Asphalt highway pavement maintenance decision method, device, equipment and storage medium |
| CN117575878A (en) * | 2023-11-16 | 2024-02-20 | 杭州众诚咨询监理有限公司 | Intelligent management method and device for traffic facility asset data, electronic equipment and medium |
| CN117575878B (en) * | 2023-11-16 | 2024-04-26 | 杭州众诚咨询监理有限公司 | Intelligent management method and device for traffic facility asset data, electronic equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113963285B (en) | 2022-06-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113963285B (en) | Road maintenance method and equipment based on 5G | |
| US11887064B2 (en) | Deep learning-based system and method for automatically determining degree of damage to each area of vehicle | |
| CN110197203B (en) | Bridge pavement crack classification and identification method based on width learning neural network | |
| CN110991466A (en) | Highway road surface condition detecting system based on novel vision sensing equipment | |
| CN110969166A (en) | Small target identification method and system in inspection scene | |
| CN110569856B (en) | Sample labeling method and device, and damage category identification method and device | |
| CN110942050A (en) | Automatic vending machine commodity identification system based on image processing | |
| CN111462140B (en) | Real-time image instance segmentation method based on block stitching | |
| CN114926766A (en) | Identification method and device, equipment and computer readable storage medium | |
| WO2021190269A1 (en) | Vehicle loss assessment method, vehicle loss assessment apparatus, and electronic device using same | |
| Hascoet et al. | Fasterrcnn monitoring of road damages: Competition and deployment | |
| JP2020038132A (en) | Crack on concrete surface specification method, crack specification device, and crack specification system, and program | |
| CN110738576B (en) | Method and device for generating damage assessment file for damaged vehicle | |
| CN115272222A (en) | Method, device and equipment for processing road detection information and storage medium | |
| CN117078677B (en) | Defect detection method and system for starting sheet | |
| CN103824074A (en) | Crowd density estimation method based on background subtraction and texture features and system | |
| Katsamenis et al. | A Few-Shot Attention Recurrent Residual U-Net for Crack Segmentation | |
| CN116343100A (en) | Target identification method and system based on self-supervision learning | |
| CN111311601A (en) | Segmentation method and device for spliced image | |
| CN110660000A (en) | Data prediction method, device, equipment and computer readable storage medium | |
| JP2019211921A (en) | Object recognition system and object recognition method | |
| Wang et al. | Detection of distress region from subway tunnel images via U-net-based deep semantic segmentation | |
| CN114998713A (en) | Pavement disease identification method, device, system, electronic equipment and storage medium | |
| CN113421319A (en) | Image processing method and device and computer equipment | |
| CN114596439A (en) | Image-based intelligent damage assessment and check method and device for car insurance and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 250101 no.1188 Tianchen street, high tech Zone, Jinan City, Shandong Province Applicant after: Shandong Jinyu Information Technology Group Co.,Ltd. Address before: 250101 no.1188 Tianchen street, high tech Zone, Jinan City, Shandong Province Applicant before: JINAN JINYU HIGHWAY INDUSTRY DEVELOPMENT Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A road maintenance method and equipment based on 5G Effective date of registration: 20230113 Granted publication date: 20220610 Pledgee: Ji'nan rural commercial bank Limited by Share Ltd. high tech branch Pledgor: Shandong Jinyu Information Technology Group Co.,Ltd. Registration number: Y2023980030827 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20231222 Granted publication date: 20220610 Pledgee: Ji'nan rural commercial bank Limited by Share Ltd. high tech branch Pledgor: Shandong Jinyu Information Technology Group Co.,Ltd. Registration number: Y2023980030827 |