CN111666833A - Road damage detection method based on vehicle formation in intelligent networking environment - Google Patents
Road damage detection method based on vehicle formation in intelligent networking environment Download PDFInfo
- Publication number
- CN111666833A CN111666833A CN202010427274.4A CN202010427274A CN111666833A CN 111666833 A CN111666833 A CN 111666833A CN 202010427274 A CN202010427274 A CN 202010427274A CN 111666833 A CN111666833 A CN 111666833A
- Authority
- CN
- China
- Prior art keywords
- road
- model
- damage
- road damage
- networking environment
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 230000015572 biosynthetic process Effects 0.000 title claims abstract description 17
- 230000006855 networking Effects 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000012360 testing method Methods 0.000 claims abstract description 10
- 238000010845 search algorithm Methods 0.000 claims abstract description 7
- 230000010354 integration Effects 0.000 abstract description 3
- 238000012423 maintenance Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a road damage detection method based on vehicle formation in an intelligent networking environment, which comprises the following steps: s1: manually collecting a road damage picture and a corresponding non-damage picture; s2: taking 70% of data in the data set as a training sample set, taking the rest data as a test sample, and optimizing model parameters by adopting a grid search algorithm; s3: implanting the trained model into a vehicle navigation system, mounting road pavement image acquisition equipment on a vehicle, inputting the acquired image into the model, and outputting a result; s4: and (4) integration of results: the road management center acquires output results of all vehicles of a fleet in real time; the method can accurately identify the road damage, provides reference for road operation and maintenance, and can save the detection cost.
Description
Technical Field
The invention relates to the field of intelligent transportation, in particular to a road damage detection method based on vehicle formation in an intelligent networking environment.
Background
At present, the mileage of the highway in China exceeds 14 kilometers, and the highway is the first place in the world. With the increasing of the construction period, various diseases begin to appear on the expressway, and road damage of different degrees is caused, so that the driving safety and comfort of a driver are directly influenced. In order to discover road damage in time and repair the road damage in time, the good service performance of the road is ensured. The traditional manual maintenance method needs a management department to invest a large amount of manpower, financial resources and material resources, and certain deviation exists in detection precision.
With the continuous development of IT technology and communication technology, the automatic detection technology is applied to road damage detection, and many research institutions and enterprises develop their own detection vehicles and are commercially applied. However, the professional detection vehicle is expensive and has low detection efficiency, and the normal operation of road traffic is affected particularly on a road section with large traffic volume. In general, road damage detection is very important, and the current detection technology cannot meet the requirement of real-time and automatic high-efficiency detection. Therefore, it is necessary to provide an automatic and efficient road damage detection method in combination with the emerging technology.
Disclosure of Invention
The invention provides a road damage detection method based on vehicle formation in an intelligent networking environment, which utilizes a motorcade formed by vehicles driving for traveling to realize real-time, automatic and high-efficiency detection of road damage.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a road damage detection method based on vehicle formation in an intelligent networking environment comprises the following steps:
s1: collecting a road damage picture and a corresponding non-damage picture, importing the road damage picture and the non-damage picture into a database in a digital form to be stored to form a data set, marking the road damage picture as 1 and marking the corresponding non-damage picture as 0;
s2: using a part of data in the data set of S1 as a training sample set, using the rest data as a test sample set, and performing model training by using the training sample set and the test sample set to obtain a trained model, wherein the model is used for identifying road damage;
s3: implanting the trained model into a vehicle navigation system, installing road pavement image acquisition equipment on a motorcade vehicle, acquiring a road pavement image of the vehicle, inputting the acquired road pavement image into the model, and outputting a recognition result of 0 or 1 by the model, wherein 0 corresponds to non-damage and 1 corresponds to damage;
s4: obtaining output identification results of models of all vehicles of a fleet in real time, and if the number of output 1 is larger than or equal to the number of output 0, then the road pavement of the output result is damaged; if the number of the output 1 is smaller than that of the output 0, the output result is normal.
Preferably, the ratio of the road damage picture to the corresponding non-damage picture in step S1 is 1: 1.
Preferably, the road damage in the damage picture in step S1 includes, but is not limited to, cracks, ruts, and other road deformations.
Preferably, in step S1, a manual collection method is used to collect the road damage picture and the corresponding non-damage picture, where the manual collection method includes mobile phone photographing and camera photographing.
Preferably, the database in step S1 is one of EXCEL, Access, Oracle, MySQL, or other databases.
Preferably, in step S2, 70% of the data in the data set of S1 is used as a training sample set, and the ratio of the damaged pictures to the corresponding non-damaged pictures in the training sample set is 1: 1.
Preferably, in the step S2, in the model training using the training sample set and the testing sample set, the model hyper-parameter is optimized by using a grid search algorithm.
Preferably, the grid search algorithm is adopted to optimize the model hyper-parameters in step S2, specifically:
s2.1: setting the interval of the ith hyper-parameter needing to be optimized of the model as [ a ]i,bi]At an interval of ci；
S2.2: keeping other super parameter values of the model unchanged, adjusting the ith (i is 1, …, N) super parameter, increasing c each time, and adjusting in total
Then, obtain
(ii) a result;
s2.3: repeating the step S2.1 and the step S2.2 to adjust other hyper-parameters of the model until N parameters in the model are adjusted;
s2.4: obtained at last
And selecting the optimal value from the results, wherein the corresponding parameter combination is the optimal parameter of the model.
Preferably, the vehicle navigation system in step S3 is one or both of a mobile phone navigation system and a vehicle navigation system.
Preferably, the road surface image acquisition device in the step S3 includes a camera.
Preferably, the fleet of vehicles is a number of vehicles passing through a section of the road continuously over a period of time, the number of vehicles being greater than 1.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
1. the invention widely uses the technology of the Internet of things, and makes up the defects of high cost and low automation degree of the traditional manual detection;
2. the invention provides an integration method, wherein a novel model and a novel model for future research can be applied in the integration method;
3. the invention adopts the idea of integrated learning in machine learning and combines the idea of edge calculation, thereby improving the accuracy of the detection result and providing reliable decision basis for traffic planning and operation management departments.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
FIG. 2 is a schematic diagram of a model parameter optimization process in the embodiment.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Example 1
The embodiment provides a road damage detection method based on vehicle formation in an intelligent networking environment, as shown in fig. 1, including the following steps:
s1: collecting a road damage picture and a corresponding non-damage picture, importing the road damage picture and the non-damage picture into a database in a digital form to be stored to form a data set, marking the road damage picture as 1 and marking the corresponding non-damage picture as 0;
s2: using a part of data in the data set of S1 as a training sample set, using the rest data as a test sample set, and performing model training by using the training sample set and the test sample set to obtain a trained model, wherein the model is used for identifying road damage;
s3: implanting the trained model into a vehicle navigation system, installing road pavement image acquisition equipment on a motorcade vehicle, acquiring a road pavement image of the vehicle, inputting the acquired road pavement image into the model, and outputting a recognition result of 0 or 1 by the model, wherein 0 corresponds to non-damage and 1 corresponds to damage;
s4: obtaining output identification results of models of all vehicles of a fleet in real time, and if the number of output 1 is larger than or equal to the number of output 0, then the road pavement of the output result is damaged; if the number of the output 1 is smaller than that of the output 0, the output result is normal.
The ratio of the road damage picture to the corresponding non-damage picture in step S1 is 1: 1.
The road damage in the damage picture in step S1 includes, but is not limited to, cracks, ruts, and other road deformations.
In the step S1, a manual collection method is adopted to collect the road damage picture and the corresponding non-damage picture, wherein the manual collection method includes mobile phone photographing and camera photographing.
The database in step S1 is one of EXCEL, Access, Oracle, and MySQL, or other databases.
In step S2, 70% of the data in the data set of S1 is used as a training sample set, and the ratio of the damaged pictures to the corresponding non-damaged pictures in the training sample set is 1: 1.
In the step S2, in the model training using the training sample set and the test sample set, a grid search algorithm is used to optimize the model hyper-parameters.
In step S2, a grid search algorithm is used to optimize the model hyper-parameters, as shown in fig. 2, specifically:
s2.1: setting the interval of the ith hyper-parameter needing to be optimized of the model as [ a ]i,bi]At an interval of ci；
S2.2: keeping other super parameter values of the model unchanged, adjusting the ith (i is 1, …, N) super parameter, increasing c each time, and adjusting in total
Then, obtain
(ii) a result;
s2.3: repeating the step S2.1 and the step S2.2 to adjust other hyper-parameters of the model until N parameters in the model are adjusted;
s2.4: obtained at last
And selecting the optimal value from the results, wherein the corresponding parameter combination is the optimal parameter of the model.
In step S3, the vehicle navigation system is one or both of a mobile phone navigation system and a vehicle navigation system.
The road surface image acquisition device in the step S3 includes a camera and a camera.
The fleet is a plurality of vehicles which pass through a road section for a period of time continuously, and the number of the vehicles is more than 1.
The same or similar reference numerals correspond to the same or similar parts;
the terms describing positional relationships in the drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. A road damage detection method based on vehicle formation in an intelligent networking environment is characterized by comprising the following steps:
s1: collecting a road damage picture and a corresponding non-damage picture, importing the road damage picture and the non-damage picture into a database in a digital form to be stored to form a data set, marking the road damage picture as 1 and marking the corresponding non-damage picture as 0;
s2: using a part of data in the data set of S1 as a training sample set, using the rest data as a test sample set, and performing model training by using the training sample set and the test sample set to obtain a trained model, wherein the model is used for identifying road damage;
s3: implanting the trained model into a vehicle navigation system, installing road pavement image acquisition equipment on a motorcade vehicle, acquiring a road pavement image of the vehicle, inputting the acquired road pavement image into the model, and outputting a recognition result of 0 or 1 by the model, wherein 0 corresponds to non-damage and 1 corresponds to damage;
s4: obtaining output identification results of models of all vehicles of a fleet in real time, and if the number of output 1 is larger than or equal to the number of output 0, then the road pavement of the output result is damaged; if the number of the output 1 is smaller than that of the output 0, the output result is normal.
2. The method for detecting road damage based on vehicle formation in the intelligent networking environment according to claim 1, wherein the ratio of the road damage picture to the corresponding non-damage picture in step S1 is 1: 1.
3. The method for detecting road damage based on vehicle formation in the intelligent networking environment according to claim 2, wherein the road damage in the damage picture in the step S1 includes cracks and ruts.
4. The road damage detection method based on vehicle formation in the intelligent networking environment according to claim 3, wherein in step S1, a manual acquisition method is adopted to acquire the road damage picture and the corresponding non-damage picture, and the manual acquisition method comprises mobile phone photographing and camera photographing.
5. The method for detecting road damage based on vehicle formation in the intelligent networking environment according to claim 4, wherein the database in the step S1 is one of EXCEL, Access, Oracle and MySQL.
6. The method for detecting road damage based on vehicle formation in the intelligent networking environment according to any one of claims 1 to 5, wherein 70% of data in the data set of S1 is used as a training sample set in step S2.
7. The method for detecting road damage based on vehicle formation in the intelligent networking environment of claim 6, wherein in the step S2, a grid search algorithm is adopted to optimize the model hyper-parameters during model training by using the training sample set and the test sample set.
8. The method for detecting road damage based on vehicle formation in the intelligent networking environment according to claim 7, wherein the step S2 is implemented by optimizing model hyper-parameters by using a grid search algorithm, specifically:
s2.1: setting the interval of the ith hyper-parameter needing to be optimized of the model as [ a ]i,bi]At an interval of ci；
S2.2: keeping other super parameter values of the model unchanged, adjusting the ith (i is 1, …, N) super parameter, increasing c each time, and adjusting in total
Then, obtain
(ii) a result;
s2.3: repeating the step S2.1 and the step S2.2 to adjust other hyper-parameters of the model until N parameters in the model are adjusted;
s2.4: obtained at last
And selecting the optimal value from the results, wherein the corresponding parameter combination is the optimal parameter of the model.
9. The method for detecting road damage based on vehicle formation in the intelligent networking environment according to claim 7 or 8, wherein the vehicle navigation system in the step S3 is one or both of a mobile phone navigation system and a vehicle navigation system.
10. The method for detecting road damage based on vehicle formation in the intelligent networking environment according to claim 9, wherein the road surface image acquisition device in the step S3 comprises a camera and a camera.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2020100642649 | 2020-01-20 | ||
| CN202010064264 | 2020-01-20 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111666833A true CN111666833A (en) | 2020-09-15 |
Family
ID=72383948
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010427274.4A Pending CN111666833A (en) | 2020-01-20 | 2020-05-19 | Road damage detection method based on vehicle formation in intelligent networking environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111666833A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815912A (en) * | 2019-01-28 | 2019-05-28 | 象谱信息产业有限公司 | A kind of expressway safety inspection system based on artificial intelligence |
| CN110008848A (en) * | 2019-03-13 | 2019-07-12 | 华南理工大学 | A kind of travelable area recognizing method of the road based on binocular stereo vision |
-
2020
- 2020-05-19 CN CN202010427274.4A patent/CN111666833A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815912A (en) * | 2019-01-28 | 2019-05-28 | 象谱信息产业有限公司 | A kind of expressway safety inspection system based on artificial intelligence |
| CN110008848A (en) * | 2019-03-13 | 2019-07-12 | 华南理工大学 | A kind of travelable area recognizing method of the road based on binocular stereo vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Eisenbach et al. | How to get pavement distress detection ready for deep learning? A systematic approach | |
| CN111553236B (en) | Road foreground image-based pavement disease target detection and example segmentation method | |
| CN108509954A (en) | A kind of more car plate dynamic identifying methods of real-time traffic scene | |
| CN112966941B (en) | Accident black spot identification method and system based on traffic accident big data | |
| CN113918538B (en) | New road maintenance data migration system based on artificial neural network | |
| CN110782125A (en) | Road safety risk degree evaluation method for automatically driving automobile | |
| CN114976306A (en) | Intelligent echelon utilization method of power lithium battery based on vehicle-connected cloud | |
| CN117351702A (en) | Intelligent traffic management method based on adjustment of traffic flow | |
| CN116543303A (en) | Bridge plate type rubber support disease identification and detection method based on deep learning | |
| CN114419421A (en) | Subway tunnel crack identification system and method based on images | |
| CN116109616A (en) | Pavement crack detection and small-surface element fitting detection method based on YOLOv5 | |
| CN117152513A (en) | Vehicle boundary positioning method for night scene | |
| CN115662122A (en) | GPS-based road inspection, disease identification and inspection and examination maintenance system and method | |
| CN114971166A (en) | Lightweight road pavement service quality inspection system | |
| CN114387261A (en) | Automatic detection method suitable for railway steel bridge bolt diseases | |
| CN117788950A (en) | Pavement disease detection method based on improved YOLOv8 model | |
| CN117197085A (en) | Road rapid-inspection image pavement disease detection method based on improved YOLOv8 network | |
| CN117670855A (en) | RoadU-Net-based intelligent recognition and classification method for asphalt pavement diseases | |
| CN116718255A (en) | Intelligent earth and stone side traffic weighing system | |
| CN111666833A (en) | Road damage detection method based on vehicle formation in intelligent networking environment | |
| CN117058069A (en) | Automatic detection method for apparent diseases of pavement in panoramic image | |
| CN110765900A (en) | DSSD-based automatic illegal building detection method and system | |
| CN116363426A (en) | Automatic detection method for health state of highway antiglare shield | |
| CN116052110A (en) | Intelligent positioning method and system for pavement marking defects | |
| CN115170783A (en) | Expressway pavement crack detection method using street view image |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200915 |